NURS 521 Week 4 Informatics Paper Part 1: Assessment of a Critical Situation Example

Informatics Paper Part 1: A Tenfold Insulin Infusion Overdose Programmed Outside the Drug Library on a Step-Down Unit

Of the three critical situations offered in this assignment, this paper assesses the one involving an intravenous insulin overdose, because it brings together the technologies at the center of inpatient medication safety: the smart infusion pump, the electronic health record and bar-code scanning. It also illustrates a theme of the course readings, that health information technology improves safety only when it is designed, configured and used as a whole system (Nelson & Staggers, 2018). The paper describes the event, assesses its causes with a sociotechnical framework, reviews research on the informatics solutions involved, and sets out findings that will guide the implementation policy in Part 2.

The Critical Situation

The setting is a 24-bed cardiac step-down unit in a 320-bed community hospital. At 0245, a 71-year-old patient with type 2 diabetes, admitted with a heart failure exacerbation and a glucose of 412 mg/dL, was started on an intravenous regular insulin infusion at a standard concentration of 1 unit per milliliter, ordered at 5 units per hour. The night nurse, caring for five patients and covering a colleague's break, scanned the patient's wristband and the insulin bag with the bar-code scanner, which matched the order. The nurse then programmed the pump. The drug library entry for insulin required selecting the concentration and entering a dose in units per hour; the nurse, who had found the library menu slow, selected basic infusion mode instead and entered a rate of 50 mL per hour, intending 5.

Basic mode applies no dose limits, so the pump gave no alert. The unit's policy required an independent double check for insulin infusions, but no second nurse was free, and the check was documented as completed by the charge nurse at 0310 without the pump screen being reviewed. At 0420, the patient was found diaphoretic and confused, with a point-of-care glucose of 34 mg/dL. The infusion was stopped, dextrose 50% was given, and the patient recovered fully by morning. The event was reported as a serious medication error with temporary harm.

Assessment Using a Sociotechnical Framework

A first reading of this event points to one nurse's mistake. That reading is incomplete. Sittig and Singh (2010) proposed that safety problems with health information technology arise from the interaction of eight dimensions, which run from the hardware, software and clinical content of a system, through its interface and the people who use it, to workflow, the organization's own policies and culture, outside regulation, and the way the system is measured and monitored. Assessing the event across those dimensions shows how many defenses failed together.

Hardware and Software

The pumps were smart pumps with dose error reduction software, but they were not connected to the electronic health record. Every infusion had to be typed by hand, which meant the bar-code scan verified the bag but not the numbers entered into the pump. In other words, the hospital's two main medication safety technologies did not talk to each other at the moment that mattered most.

Clinical Content

The insulin entry in the drug library had sensible limits, a soft limit at 15 units per hour and a hard limit at 25, which would have stopped this error at 50 units per hour. The content was right; it simply was not used.

Human-Computer Interface

The nurse described the library menu as slow, requiring the user to scroll through more than a hundred drug names on a small screen. Basic mode was one button away. When the safe path takes more steps than the unsafe one, the interface itself invites the workaround.

People

The nurse was experienced and competent, but working at 0300 with five patients and a colleague's break, conditions known to raise error risk. The charge nurse documenting the double check was also stretched thin, responsible for a rapid response call two rooms away at the same time. Neither nurse had received refresher training on the pump since it was installed three years earlier, and newer staff reported learning to use basic mode from colleagues during orientation rather than from the hospital's own training. The people dimension, then, includes not only fatigue but a gap in how the hospital teaches and refreshes the use of its safety technology.

Workflow and Communication

The independent double check, the last human defense, had become a signature rather than a verification. No process required the second nurse to read the pump screen against the order. The record offered a double-check field that accepted any nurse's name, so the electronic documentation gave the appearance of a safeguard that had not taken place. Communication failed as well: the bedside glucose checks were ordered hourly, but the first check after the infusion began was not due until 0345 and was delayed by the rapid response call, so the falling glucose went unnoticed for more than an hour.

Internal Organizational Policies, Procedures and Culture

Hospital policy discouraged basic mode but did not prohibit it, and basic mode use was not reviewed. Staff described double-check documentation without a real check as common at night.

External Rules and Regulations

Federal rules do not require pump interoperability, and the Institute for Safe Medication Practices sets national guidance on smart pump use, including goals for drug library compliance, rather than mandatory standards.

System Measurement and Monitoring

The pump vendor's software logged every infusion, including basic mode use, but no one reviewed the reports. Drug library compliance on the unit, calculated after the event, was 71%, meaning nearly three in ten infusions ran without dose limits.

The Role of Informatics in the Situation

Informatics is central to both the cause and the solution. The event happened in the gap between two information systems, the electronic record that held a correct order and the pump that received an incorrect number typed by hand. The data needed to see the risk in advance, the unit's low library compliance, already existed in the pump logs but was never turned into information anyone used. Nelson and Staggers (2018) describe informatics as the work of transforming data into information and knowledge that support decisions, and this hospital had the data but not the transformation. The research reviewed below suggests that informatics tools can close both gaps: interoperability to remove manual programming, and analytics to make compliance visible.

Research Supporting the Assessment

Interoperability Reduces Programming Errors

Borrelli et al. (2025) systematically reviewed studies of medication administration errors before and after smart pump interoperability in hospitals. Three studies met their criteria, covering community hospitals, pediatric facilities and intensive care units. After implementation, error types that interoperability can affect fell by 15.4% to 54.8%, and overall administration errors fell by 21.2% to 90.5%, with the size of the change depending on baseline compliance and setting. The small number of studies is a limitation, but every study pointed in the same direction.

Integration Raises Library Use and Reduces Alerts

In six hospitals of an academic health system, Afaq et al. (2026) compared infusions before and after pumps were integrated with the electronic record. After integration, hard limit alerts fell by 50.3%, soft limit alerts by 30.4% and alerts for large single-step rate changes by 38.1%. Drug library compliance rose to 96.1%. Auto-programmed infusions needed 1.6 programming steps compared with 3.3 for manual programming, and alerts were resolved faster. Notably, auto-programming itself was used for only 53.4% of eligible infusions overall and 68.3% for high-alert medications, showing that integration must be followed by training and monitoring if it is to be used.

The Wider Evidence Is Positive but Observational

Sowan et al. (2026) reviewed 20 records, 14 full studies and 6 conference proceedings, on pump-record interoperability and found that most reported positive associations with safety, efficiency, charge capture and user experience. They cautioned that most studies were observational and inconsistently reported, and they concluded that interoperability should be treated as one component of a sociotechnical system in which users, workflow, clinical content and organizational practices together decide whether it works. That conclusion matches the framework used in this assessment.

Analytics Make Compliance Visible

Odinet et al. (2026) implemented an infusion analytics platform across a 47-hospital health system and analyzed more than 5.6 million infusions over 18 months. Total drug library alerts fell from 28,040 to 10,960, interoperability compliance rose modestly from 67.4% to 68.9%, and drug library managers reported that monthly compliance reporting took about half the time it had, with event investigations taking about three quarters less time. The study shows that the pump logs the hospital in this situation ignored can be turned into routine, actionable reports.

Findings

The assessment leads to five findings that Part 2 will address:

• The pumps and the electronic record are not interoperable, so every infusion depends on manual programming.
• Basic mode is permitted for high-alert medications and is easier to reach than the drug library.
• Drug library compliance on the unit is 71%, and no one monitors it.
• The independent double check for high-alert infusions does not reliably include verifying the pump screen.
• Night staffing leaves little capacity for double checks and careful programming.

Conclusion

A tenfold insulin overdose on this unit was not the act of a careless nurse but the predictable result of systems that left a correct order and a wrong pump setting unconnected. A sociotechnical assessment shows failures in the technology, the interface, the workflow, the policies and the monitoring, and current research shows that interoperability and compliance analytics can address several of them at once. Part 2 will turn these findings into an implementation policy for safe programming of high-alert infusions.

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

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