| Course | NU710 Advanced Health Information and Technologies |
|---|---|
| Unit | Unit 2 |
| Paper type | Clinical decision support analysis |
| Length | About 1,110 words, 7 pages double-spaced |
| Format | APA 7 student paper |
| School | Purdue University Global |
| Program | Doctor of Nursing Practice (DNP) |
| Updated | October 2026 |
Free sample paper for NU710 Unit 2
Fourteen Thousand Glucose Alerts a Month: A Unit 2 Clinical Decision Support Analysis of an Insulin Alert Nurses Learned to Ignore
[Student Name]
Purdue University Global
NU710: Advanced Health Information and Technologies
Unit 2 Clinical Decision Support Analysis
[Instructor Name]
[Date]
The hospital, its alerts, figures and nurses are composite and illustrative; the studies cited are real.
The Alert and Its Purpose
The alert was built in 2019 after a review found that correction insulin was sometimes missed when glucose was high before a scheduled dose. It fires when a nurse opens the medication administration record for a patient with a scheduled insulin dose and a point-of-care glucose above 180 mg/dL in the prior two hours. It interrupts the nurse with a box that reads "Hyperglycemia: review correction scale," and requires a click to continue.
The intention was sound. In practice, the alert fires for nearly every patient with diabetes before most meals, because a glucose above 180 before a meal is common and the correction scale is already part of the order. The nurse who sees the alert has usually already read the glucose and calculated the correction.
What the Data Show
A three-month extract from the electronic record, January to March, shows the scale of the problem. The alert fired 42,300 times, about 14,100 a month, across ten inpatient units. Nurses took a documented action, a change to the dose or a provider notification, after 2,560 firings, about 6 percent. On the two pilot units of the insulin discharge project, nurses saw the alert a median of 11 times per shift. At an average of six seconds per firing, the alert occupied about 70 nursing hours a month across the hospital.
Thompson et al. (2024) faced a similar problem at a children's hospital, where four alerts made up 43 percent of interruptive nursing alerts and an estimated 86 nursing hours a month. A small team redesigned them using the five rights of decision support and usability testing, and interruptive nursing alert firings fell by 58 percent while the share of firings that led to action rose from 8.1 to 17.3 percent.
An alert that is right 6 percent of the time teaches nurses that it is wrong.
Fit Between Individuals, Tasks and Technology
Berkhout et al. (2025) interviewed 22 hospital nurses about decision support and organized barriers and facilitators using the FITT framework, which looks at the fit between the individual user, the task and the technology. Their participants described alarm fatigue, poor usability, weak integration with workflow and limited training as barriers. Applying the same three-way fit to this alert explains its failure.
Individual and task. Experienced nurses already check glucose and calculate correction doses before opening the medication record. The alert adds nothing to their knowledge and interrupts a task they have already done.
Task and technology. The alert fires at the wrong moment in the workflow, after the decision has been made, and offers no shortcut to act. A nurse who does need to change a dose must close the alert and click through two more screens to reach the order.
Individual and technology. Newer nurses, the group the alert would most help, say they stopped reading it within weeks because it fired constantly. In a short survey on the pilot units, 31 of 38 nurses said they clicked through without reading it most of the time.
Where the Alert Breaks the Five Rights
Measured against the five rights of decision support, the current alert scores badly on nearly every one. The information is generic rather than patient-specific. It reaches every nurse, not the one at risk of missing a dose. Its format is an interruptive box. It arrives after the decision.
Proposed Redesign
The redesign narrows the alert to the situations where it adds value and moves the rest to passive display.
1. Narrow the trigger. Fire the interruptive alert only when a correction dose is due and has not been documented 30 minutes after a meal tray is recorded, or when glucose is above 300.
2. Change the format for other cases. Show glucose and the calculated correction dose in a non-interruptive banner on the medication record, so the information is present when the nurse is deciding.
3. Add an action. When the alert does fire, include a button that opens the correction dose order directly.
4. Add a hypoglycemia rule. Replace part of the volume with a rule the hospital lacks: a non-interruptive flag when a scheduled basal dose is due and the most recent glucose is below 100, which is more relevant to the discharge project's concern with lows.
The redesign will be tested with six nurses from three units in a simulated medication pass before go-live, following Thompson et al.'s approach of in situ usability testing.
Measures
The redesign will be judged on four measures, compared for three months before and after: interruptive firings per month, the share of firings followed by action, the rate of missed correction doses found on chart audit and, as a balancing measure, any glucose under 70 recorded in the six hours after a correction dose was given. Nurse perception will be measured with the same three-question survey.
Why Nurses Matter Most Here
Most published work on alert fatigue concerns physicians and prescribing alerts. Nurses meet a different set of alerts, at the moment of administration rather than ordering, and often under more time pressure, since a medication pass for five patients cannot wait while each alert is read. Berkhout et al. (2025) note that nursing use of decision support is less studied than physician use, which helps explain why this alert was never reviewed: nobody owned nursing alerts. A redesign led by nurses, tested with nurses and measured on nursing workflow is the correction.
Governance
The hospital has no standing committee for nursing alerts, so nobody watches the measurement and monitoring dimension that a sociotechnical view of safety treats as part of the system itself (Sittig & Singh, 2010). The proposal asks the nursing informatics council to own nursing decision support, to review every interruptive nursing alert yearly using firing and action data, and to approve new alerts only with a stated purpose and a measure of success.
Conclusion
The hyperglycemia alert was designed to prevent missed insulin and instead taught nurses to click past it. Its failure is a failure of fit with nursing work and of nearly every one of the five rights. Narrowing the trigger, moving information to the point of decision, adding a direct action and governing nursing alerts with data would reduce the burden and make the remaining alerts worth reading.
References
Berkhout, M., Smit, K., Sent, D., Kusters, R., Versendaal, J., & van Houwelingen, T. (2025). Understanding the role of clinical decision support systems among hospital nurses using the FITT (Fit Between Individuals, Tasks, and Technology) framework: Qualitative study. Journal of Medical Internet Research, 27, e76025. https://doi.org/10.2196/76025
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
Thompson, S. A., Kandaswamy, S., & Orenstein, E. (2024). A discount approach to reducing nursing alert burden. Applied Clinical Informatics, 15(4), 727-732. https://doi.org/10.1055/a-2345-6475
Reading the NU 710 Unit 2 instructions
Unit 2 asks students to examine a decision support tool or alert that nurses meet, and many prompts specifically want data on how it performs. Expect to report how often it fires, what share of firings lead to action, how it fits nurses' workflow and what should change. A framework such as FITT or the five rights gives the analysis structure. Instructors look for a redesign that follows from the analysis, with each change answering a specific failure, and for measures to judge it, including a balancing measure in case the change causes harm. Published redesigns are useful as benchmarks. The best papers also ask who governs nursing alerts, since alerts built without an owner tend to stay in place long after they stop helping anyone.
How this NU710 Unit 2 example is built
The paper starts with the alert's history and the sound reason it was built, then shows that it fires before nearly every meal for patients with diabetes. A data section reports firings, action rate, firings per shift and nursing hours, then sets Thompson and colleagues' redesign beside them under a highlighted line about an alert right 6 percent of the time. The FITT section examines fit between individual, task and technology using Berkhout's nurses and a short local survey. A section on the five rights shows where the alert breaks each. Four numbered changes follow, from a narrowed trigger to a new low-glucose flag, with usability testing. Another section explains why nursing alerts are less studied, and measures and a governance proposal close the paper.
Where the points sit in the NU710 Unit 2 rubric
Scoring tends to favor papers that put numbers on the problem: firing counts, action rates and time cost carry real weight. Application of a framework is judged on whether it explains the alert's failure in this setting. The redesign earns most when each change is tied to an identified failure and when usability testing is planned before go-live. Measures should include a balancing measure, and graders often check for one. A benchmark from published work strengthens the case. Governance, saying who owns nursing decision support, is increasingly expected at the DNP level. Writing that a nurse manager or informatics council could act on, plus scholarly sources and APA 7, completes the rubric for this analysis.
NU710 Unit 2 help from the desk
Alert analyses lose ground when they describe alert fatigue in general and never measure the alert at hand. Another frequent problem is recommending that the alert be removed without asking what risk it was built to address. Some papers propose a redesign without testing it or measuring it afterward. Leaving out a balancing measure invites the question of whether fewer alerts led to missed doses. Writers also forget nurses' workflow, proposing changes that make sense on a screen but not during a medication pass. Start by asking your informatics team for three months of firing and override data on one alert, then watch two nurses meet it. A custom analysis can work from the alert your own unit complains about most.
Get an NU710 Unit 2 example written to your instructions
Share the alert or tool, any data on how it fires and the analysis instructions with the rubric. A custom analysis examines its fit with nursing work and proposes a redesign. Turnaround is 24-48h, free the first time. The paper above is an original model document written by our desk, not a submitted student paper and not an official Purdue University Global document.
More NU710 papers and DNP samples
- NU710 Unit 1: Health IT Safety Event Post
- NU710 Unit 3: Predictive Analytics Evaluation
- NU710 Unit 4: AI Ethics Analysis
- NU710 Unit 5: Data Confidentiality Analysis
- NU710 Unit 6: Cyberattack and Downtime Response Plan
- NU710 Unit 7: Technology Implementation Plan
- NU710 Unit 8: Device Integration Evaluation
- NU710 Unit 9: Technology Impact Assessment
- NU710 Unit 10: E-health Tool Proposal
- NU713 Unit 2: Social Determinants Profile
- NU708 Unit 2: System Map Analysis
- NU733 Unit 2: Discussion Post, Costs of a Clearance
- NU706 Unit 2: Distributive Justice Analysis
NU710 Unit 2 questions, answered
What is alert fatigue?
The tendency to ignore or override alerts when too many fire, especially ones that rarely require action. It can lead clinicians to miss the important ones.
What are the five rights of clinical decision support?
Delivering the right information to the right person in the right format through the right channel at the right time in the workflow.
Should an alert analysis include data?
Yes, ideally firing counts and the share of firings that led to action, before and after any change.
Where can I find a free NU710 Unit 2 sample paper?
The complete NU710 Unit 2 clinical decision support analysis appears on this page, covering a hyperglycemia alert with a 6 percent action rate, its poor fit with nursing work and a four-part redesign with measures.
What new rule does the NU710 Unit 2 sample add?
A non-interruptive flag when a scheduled basal insulin dose is due and the latest glucose is below 100, addressing the risk of lows that matters most to the discharge project.