NUR 8150 Module 2 Implementation Risk Paper: CPOE Error Pathways, a Mortality Warning and a Safer Go-Live Plan Example

Twenty-Two Ways to Make an Error: Learning From Order Entry Failures to Plan a Safer Go-Live at a 280-Bed Hospital

Student Name

Doctor of Nursing Practice, Thomas Edison State University

NUR-8150: Information Systems and Technology Impacting Healthcare Delivery

Mentor Name

Month Day, Year

Twenty-Two Ways to Make an Error: Learning From Order Entry Failures to Plan a Safer Go-Live at a 280-Bed Hospital

The Decision Before Us

The hospital in this paper is invented: a 280-bed community hospital in Toms River with a 24-bed intensive care unit, a busy emergency department and a small pediatric unit. Its order entry system is ten years old and will lose vendor support in eighteen months. The board has approved a new integrated system. The chief nursing officer has asked the DNP-prepared director of nursing informatics to lead the clinical side of implementation and to answer one question first: what could go wrong, and how do we prevent it?

Why Adoption Was Slow

Jha et al. (2009) asked every acute care member of the American Hospital Association which electronic functions it had in place. Among the 63.1 percent that responded, only 1.5 percent had a comprehensive electronic record system in all clinical units while a further 7.6 percent ran a basic record on one or more wards. Medication order entry had been implemented in only 17 percent. Systems were commoner in big, urban and teaching hospitals, and hospitals cited the capital cost and the cost of maintenance as the main barriers.

That history matters for this hospital because many of its staff learned on the current system, which was installed when adoption was rising quickly and training was short. The new system will be a large change for nurses, pharmacists and prescribers, and the lessons from early adopters are the best guide to its risks.

How Order Entry Creates Errors

Koppel et al. (2005) examined one hospital's order entry product through a survey answered by 261 residents, focus groups, interviews and direct observation. In all, the product opened 22 separate routes to a medication error. Examples included screens that showed only part of a patient's medicines at once, so no one saw the whole list; pharmacy inventory displays that prescribers mistook for dosing guidance; reminders to renew antibiotics that lived on paper while orders lived on screens; separate functions that allowed double doses and conflicting orders; and rigid ordering formats that produced wrong orders. About three quarters of house staff had seen each risk, many weekly or more often.

A Warning About Speed

Han et al. (2005) tracked deaths among 1,942 children flown or driven in from other hospitals across 18 months that included a six-day, all-units switch to a commercial order entry product. Mortality rose from 2.80 percent in the 13 months before to 6.57 percent in the five months after, and the system remained linked with higher odds of death after adjusting for other factors. The authors described delays at the bedside: orders could not be entered until the patient was registered, medicines were held in a central pharmacy rather than on the unit, and nurses and physicians spent time at computers rather than with critically ill children.

The paper does not suggest that order entry is dangerous in itself. Other hospitals using the same product did not report the same result, and the study was observational. But the findings point to the risks of a rapid rollout in critical care, of workflows that delay urgent medicines and of too little attention to how clinicians actually work.

The Go-Live Plan

The hospital's plan is built around those lessons.

First, a phased rollout. Medical-surgical units go live first, then the emergency department, then the intensive care unit and pediatrics last, at least six weeks after the first phase, with time to fix problems.

Second, workflow testing. Clinicians from each unit walk through their most urgent scenarios, such as a code, a sepsis bundle and a transfer from another hospital, using the test system, and every delay is fixed before go-live. Emergency medicines remain available on units, and patients can be registered quickly so orders are never held.

Third, front-line design. Order sets are reviewed by the nurses and prescribers who will use them, and the medication display is checked to make sure a patient's full list appears on one screen.

Fourth, support at the elbow. Super-users from each unit and vendor staff are on every shift for two weeks after each phase.

Fifth, a rapid response line for system problems that affect patient care, staffed around the clock.

Measuring Safety

Before and after each phase, the hospital will track medication errors reported through its safety system, minutes from order to first dose for sepsis antibiotics, clot-busting drugs and cardiac medicines, risk-adjusted mortality and rapid response calls. A daily huddle reviews any system-related event. Stop rules are set in advance: if time to first antibiotic dose in sepsis rises by more than 30 minutes, or if a serious event is linked to the system, the next phase pauses until the cause is fixed.

Limits

The studies cited are from the early years of order entry, and systems have improved. The Koppel study was at one hospital, and the Han study was observational. The paper uses them not as predictions but as a catalog of what has gone wrong before, so this hospital can check for each problem.

Conclusion

Order entry adoption was slow for cost reasons, and early adopters learned painful lessons. A teaching hospital study cataloged 22 ways a system can create medication errors, and a children's hospital saw mortality rise after a six-day rollout. This hospital's plan responds with a phased rollout ending in the highest-risk units, workflow testing of urgent scenarios, front-line design of order sets and displays, on-site support and safety measures with stop rules, so that the new system improves safety rather than threatening it.

References

Han, Y. Y., Carcillo, J. A., Venkataraman, S. T., Clark, R. S. B., Watson, R. S., Nguyen, T. C., Bayir, H., & Orr, R. A. (2005). Unexpected increased mortality after implementation of a commercially sold computerized physician order entry system. Pediatrics, 116(6), 1506-1512. https://doi.org/10.1542/peds.2005-1287

Jha, A. K., DesRoches, C. M., Campbell, E. G., Donelan, K., Rao, S. R., Ferris, T. G., Shields, A., Rosenbaum, S., & Blumenthal, D. (2009). Use of electronic health records in U.S. hospitals. New England Journal of Medicine, 360(16), 1628-1638. https://doi.org/10.1056/NEJMsa0900592

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