NURS 521 Week 7 Informatics Paper Part 2: Implementation Policy Example

Informatics Paper Part 2: An Implementation Policy for Auto-Programmed, Drug Library Infusions of High-Alert Medications

The Part 1 assessment of a tenfold insulin overdose on a cardiac step-down unit identified five problems: smart pumps not connected to the electronic health record, basic mode permitted for high-alert medications, unmonitored drug library compliance of 71%, a double check that did not include reading the pump, and night staffing that left little capacity for checks. This paper answers those findings with a plan of action written as a hospital implementation policy. The policy uses three informatics applications: smart pump and record interoperability for auto-programming, drug library configuration that removes basic mode for high-alert drugs, and an infusion analytics dashboard for monitoring. It is written in the care delivery format the hospital uses for nursing policies, followed by the evidence behind it and a plan for rollout.

Policy

Policy Header

Policy NUR-MED-214: Programming and Verification of High-Alert Medication Infusions. Applies to: all inpatient nursing units, procedural areas and the emergency department. Approved by: Chief Nursing Officer, Director of Pharmacy and the Medication Safety Committee. Effective date: the first day of Phase 2 of the rollout. Review: every two years or after any serious infusion-related event.

Purpose

To prevent programming errors in continuous infusions of high-alert medications by requiring that these infusions be auto-programmed from the electronic health record through the pump's drug library, that pump settings be independently verified, and that compliance be monitored and reported.

Scope

This policy applies to every registered nurse, nursing student under supervision, pharmacist and provider involved in ordering, preparing or administering intravenous infusions of medications on the hospital's high-alert list, including insulin, heparin, opioids, sedatives, vasoactive agents, potassium and magnesium concentrates, and neuromuscular blockers.

Definitions

• Auto-programming: transfer of verified infusion parameters from the electronic health record to the smart pump, started when the nurse scans the wristband, the drug label and the pump channel, after which the nurse confirms the settings.
• Drug library: the pharmacy-built list of medications, concentrations and dosing limits stored in the pump, also called dose error reduction software.
• Basic mode: programming the pump by rate or volume without using the drug library, so no limits apply.
• Independent double check: verification by a second qualified clinician who reads the order, the medication label and the pump screen separately, without being told the expected values, before the infusion starts.

Policy Statements

• 1. Infusions of high-alert medications must be auto-programmed when the medication and pump are interoperable.
• 2. When auto-programming is unavailable, high-alert infusions must be programmed manually through the drug library. Basic mode is disabled for high-alert entries and must not be used for these medications.
• 3. An independent double check of the pump screen is required at initiation, at any bag or syringe change, at every rate change for insulin and heparin, and at handoff.
• 4. Overrides of soft limits must be documented with a reason in the record, and hard limits must not be bypassed.
• 5. Pump compliance data are reviewed monthly by unit leaders and quarterly by the Medication Safety Committee.

Procedure for the Bedside Nurse

• 1. Review the active order in the electronic record and confirm the patient's latest relevant laboratory value, such as glucose for insulin or anti-Xa level for heparin.
• 2. Scan the patient's wristband, the medication label and the barcode on the pump channel to launch auto-programming.
• 3. Compare the parameters displayed on the pump with the order, including drug, concentration, dose, dose unit and rate. Do not press start.
• 4. Ask a second registered nurse to perform the independent double check by reading the order, the label and the pump screen without prompting. Both nurses document the check in the record, which now requires the second nurse to scan their badge at the pump.
• 5. Start the infusion and confirm the pump shows the drug library name, not "basic infusion."
• 6. If auto-programming fails, program through the drug library manually, repeat step 4, and submit a pump issue report through the electronic reporting system.
• 7. For insulin infusions, check capillary glucose at the frequency in the order set, with the first check due no later than one hour after initiation.
• 8. At handoff, the receiving nurse reads the pump screen against the order with the departing nurse.

Roles and Responsibilities

The drug library manager in pharmacy maintains drug library entries, reviews soft and hard limits quarterly, and leads the review of override data. Clinical informatics maintains the interface between the record and the pumps, resolves auto-programming failures within one business day, and builds the compliance dashboard. Nurse managers review unit compliance monthly with staff and address patterns rather than individual blame, consistent with a just culture. Unit educators provide initial training and annual competency checks. Charge nurses ensure that a second nurse is available for double checks on every shift, including night shift, and may call the house supervisor for help when staffing does not allow it. Biomedical engineering keeps pump software and libraries current across all devices.

Education and Competency

Before go-live, every nurse completes a 30-minute online module and a 20-minute hands-on session at a simulated bedside, demonstrating auto-programming, manual library programming, a correct independent double check and response to a soft limit alert. Competency is rechecked annually and after any pump software update. New hires complete the same training during orientation, taught by unit educators rather than by peers, to stop the informal teaching of basic mode identified in Part 1.

Monitoring and Evaluation

Rationale From the Evidence

Each major element of the policy rests on current research. Auto-programming is the center of the policy because, in a systematic review of pre and post interoperability studies, every included study reported fewer of the programming errors that direct order transfer is designed to prevent, with reductions as large as 54.8% (Borrelli et al., 2025). Integration also changes daily work. In a six-hospital academic system, integration reduced hard limit alerts by 50.3%, raised drug library compliance to 96.1% and halved programming steps, from 3.3 to 1.6, yet barely half of the infusions that could have been auto-programmed actually were (Afaq et al., 2026). That gap is the reason the policy sets an auto-programming target separate from the library target and assigns informatics staff to fix failed transfers quickly.

The monitoring plan follows the experience of a 47-hospital system in which an infusion analytics platform halved the time needed to produce compliance reports and cut investigation time by three quarters, and total library alerts dropped by well over half during the project (Odinet et al., 2026). Data that are easy to see are more likely to be used, and Part 1 showed that this hospital already had the data but never looked at it. The double check is redesigned rather than simply repeated, because a systematic review found little consistent evidence that double checking reduces administration errors and wide variation in adherence, from 52% to 97% of administrations (Koyama et al., 2020). A double check that requires reading the pump screen independently and scanning a badge at the pump is more likely to catch an error than a signature added later. Finally, the policy treats technology, workflow, people and monitoring together because a review of interoperability studies concluded that pump-record integration works only as part of a broader sociotechnical system (Sowan et al., 2026), the same view of safety that guided the Part 1 assessment (Sittig & Singh, 2010).

Anticipated Barriers

Three barriers are likely. The first is workflow friction: nurses who have used basic mode for years will find the drug library slower at first, and some auto-programming transfers will fail during early weeks. The policy answers this with super-users on every shift and a one-business-day repair standard for interface failures, so a nurse who meets a problem has help on the spot and sees it fixed. The second barrier is staffing at night, when a second nurse may not be free for an independent double check. Charge nurses are therefore given authority to call the house supervisor, and the dashboard reports double checks by shift so leaders can see whether nights need more support. The third is alert fatigue, since drug library limits that are set too tightly produce nuisance alerts that train nurses to override. The drug library manager reviews override data quarterly and adjusts limits that fire often without catching errors, a step the analytics platform makes practical.

Implementation Timeline

The policy rolls out in three phases over nine months. In Phase 1, months 1 through 3, pharmacy removes basic mode from high-alert library entries, clinical informatics completes the pump-record interface for insulin, heparin and the five most frequent high-alert drugs, and the dashboard goes live so baseline data can be collected. In Phase 2, months 4 through 6, all nursing staff complete training, the policy takes effect on the cardiac step-down unit and two medical-surgical units as pilots, and super-users trained on each shift support staff at the bedside. In Phase 3, months 7 through 9, the policy expands hospital-wide after the pilot data are reviewed and the procedure is adjusted. Nelson and Staggers (2018) emphasize that clinicians' involvement in testing and adapting a system before full rollout is central to adoption, so pilot nurses review the procedure at the end of Phase 2 and their changes are incorporated before Phase 3.

Conclusion

The insulin overdose assessed in Part 1 arose from the gap between a correct order and a manually typed pump setting, together with weak checks and invisible data. This implementation policy closes that gap by sending orders from the record to the pump, removing the unsafe shortcut for high-alert medications, redesigning the double check and making compliance visible each month. Its success will be judged by measurable targets rather than by its adoption on paper, and its rules can be revised as the data show what works on the units.

References

Afaq, H., Colunga, S., Hu, L., Zhang, X., Munson, S., Norton, J., Townley, P., Meyer, J., Carpenter, L., Hoh, T., Perampaladas, K., Schuler, B., Silberhorn, R., & Alam, A. (2026). Integration of smart infusion pumps with electronic medical records improves safety and staff productivity at a large academic healthcare system. American Journal of Health-System Pharmacy. Advance online publication. https://doi.org/10.1093/ajhp/zxag187

Borrelli, E. P., Lucaci, J. D., Wilson, N. S., Taneja, A., Weiss, M., & Beer, I. (2025). Evaluating the impact of smart infusion pump interoperability on reducing medication administration errors: A systematic literature review. Medical Devices: Evidence and Research, 18, 247-260. https://doi.org/10.2147/MDER.S522534

Koyama, A. K., Maddox, C. S., Li, L., Bucknall, T., & Westbrook, J. I. (2020). Effectiveness of double checking to reduce medication administration errors: A systematic review. BMJ Quality & Safety, 29(7), 595-603. https://doi.org/10.1136/bmjqs-2019-009552

Nelson, R., & Staggers, N. (2018). Health informatics: An interprofessional approach (2nd ed.). Elsevier.

Odinet, J. S., Stansbury, R., Knight, C., Maultsby, D. W., & Wilson, N. S. (2026). Implementation of an enterprise data analytics solution: A quality improvement project to enhance infusion safety through smart pump technology compliance. Journal of Patient Safety. Advance online publication. https://doi.org/10.1097/PTS.0000000000001570

Sittig, D. F., & Singh, H. (2010). A new sociotechnical model for studying health information technology in complex adaptive healthcare systems. Quality and Safety in Health Care, 19(Suppl 3), i68-i74. https://doi.org/10.1136/qshc.2010.042085

Sowan, A., Hayward, M., Holzknecht, A., Mourad, M., Rodschat, C., & Vera, A. (2026). Associations between smart infusion pump-electronic health record interoperability and healthcare outcomes: A systematic review. Intensive and Critical Care Nursing, 97, Article 104475. https://doi.org/10.1016/j.iccn.2026.104475