Heparin infusion starts mapped before and after pump integration on one unit show fewer keystrokes and faster starts, plus new failure points, in the NU515 Unit 2 workflow analysis. Searches like "nu 515 unit 2 assignment example", "nu515 unit 2 sample" and "nu515 unit 2 example" land here.
Twenty-Three Keystrokes to Four: A Before-and-After Workflow Analysis of Starting a Weight-Based Heparin Infusion With Pump Integration
[Student Name]
Purdue University Global
NU515: Innovation and Application of Health Care Information Technology
Unit 2 Assignment
[Instructor Name]
[Date]
The hospital, unit and observed timings are composite and bracketed; the guidance and research cited are published.
Scope
One high-alert task, from the nurse's review of a new heparin order to a documented running rate, observed [ten] times before integration and [ten] times after on one 32-bed medical-surgical unit. Heparin was chosen because it is weight-based, titrated by protocol and on the national list of high-alert medications, so errors in programming carry real risk.
The Before Map: Thirteen Steps
Lanes: nurse, second nurse, pump, electronic record, pharmacy.
1. Nurse reviews the order and the protocol in the record.
2. Nurse retrieves the premixed bag from the automated cabinet.
3. Nurse scans the patient's wristband.
4. Nurse scans the bag.
5. Nurse selects heparin on the pump from the drug library.
6. Nurse enters concentration.
7. Nurse enters patient weight.
8. Nurse enters dose in units per kilogram per hour.
9. Nurse enters bolus, if ordered.
10. Second nurse independently verifies drug, concentration, weight, dose and rate against the order.
11. Nurse starts the infusion.
12. Nurse returns to the workstation and documents the rate in the flowsheet.
13. Pharmacy reviews the documented rate at the next protocol check.
Keystrokes on the pump: a median of [23].
The After Map: Ten Steps
1. Nurse reviews the order and protocol.
2. Nurse retrieves the bag.
3. Nurse scans the wristband.
4. Nurse scans the bag.
5. Nurse scans the pump channel, which links the channel to the order.
6. The record sends drug, concentration, weight, dose and rate to the pump.
7. Nurse confirms the parameters on the pump screen.
8. Second nurse independently verifies the pump against the order.
9. Nurse starts the infusion.
10. The pump sends the running rate to the flowsheet, where the nurse validates it.
Keystrokes on the pump: a median of [4].
What Changed, by Type of Step
Sorting steps into care, verification and documentation shows what integration did and did not change.
1. Transcription removed: four manual entries, steps 6 to 9 before, are replaced by parameters sent from the order. This is where programming errors most often occur, and it is the main safety gain (Institute for Safe Medication Practices, 2020).
2. Verification kept: the independent double check remains, because hospital policy still requires it for heparin, although it now compares the pump screen with the order rather than with a nurse's entries.
3. Documentation changed: the manual flowsheet entry becomes validation of a value the pump sends, which removes a delay and a second chance to mistype.
4. Step added: scanning the pump channel, which creates the link between order and channel. If the wrong channel is scanned on a multichannel pump, the correct parameters go to the wrong line.
Timing
Median time from order review to documented running rate: [7.4] minutes before, [5.2] minutes after. The largest saving was in documentation; programming time fell by about [90] seconds.
Where the New Path Breaks
In [three] of the ten after-integration starts, auto-programming failed and the nurse fell back to manual entry. Two failures were wireless connection drops at the far end of the hall; one was an order written in a dose unit the pump library did not recognize. When the new path breaks, the nurse falls back to the old manual path, but without the habits of the old path, and under the assumption that the system is doing the transcription. In one of the three fallbacks, the second nurse's double check compared the pump with the order but did not notice that the rate had been entered, not sent, because the screen looked the same.
The failure path deserves as much attention as the success path. A pharmacist-led implementation at another hospital reported improved infusion documentation and safety after interoperability (Biltoft & Finneman, 2018), and the gains here are similar, but those gains depend on how often the system works and on what nurses do when it does not.
Reading the Change as a Work System
A human factors view explains why the same technology can remove one risk and create another. SEIPS 2.0 describes work as the interaction of people, tasks, tools and technologies, the physical environment and the organization, producing processes and outcomes that feed back into the system (Holden et al., 2013). Integration changed the tool, but it also changed the task, from entering numbers to confirming them, and it depended on the environment, the wireless coverage along the hall, and on the organization, the build of order units and the pump library by two different departments. Each fallback in the observations traces to one of those components rather than to the nurse. Confirmation is a different cognitive task from entry: a nurse typing a weight must attend to it, while a nurse confirming a weight on a screen can glance and accept. The double check, designed for a world of manual entry, now asks the second nurse to confirm what a machine sent, which is easier to do superficially.
What the Nurse Experiences
Nurses interviewed after the observations described the change as a relief on most starts and a trap on a few. Most starts were faster and felt safer, and the end of retyping rates into the flowsheet was the change they valued most. But several said they no longer knew, at a glance, whether a pump had been programmed by the system or by hand, and one said she had begun to trust the pump screen "the way you trust spell-check." That comment names the risk better than any count in this analysis.
Recommendations
1. Make fallback visible: the pump screen should show clearly whether parameters were sent or entered manually, and the double check should confirm which.
2. Fix connection dead zones identified in the hall.
3. Reconcile order dose units with the pump library, starting with high-alert drugs.
4. Teach channel scanning on multichannel pumps as a distinct safety step, and repeat these observations on nights, when fewer colleagues are available to catch a fallback.
Limits
Ten observations in each condition on one unit, during day shifts, with an observer present, which may change behavior. Timing includes interruptions that vary by day. The analysis covers one drug; titrated infusions in intensive care, with frequent rate changes, would stress the system differently and deserve their own map.
References
Biltoft, J., & Finneman, L. (2018). Clinical and financial effects of smart pump-electronic medical record interoperability at a hospital in a regional health system. American Journal of Health-System Pharmacy, 75(14), 1064-1068. https://doi.org/10.2146/ajhp161058
Holden, R. J., Carayon, P., Gurses, A. P., Hoonakker, P., Hundt, A. S., Ozok, A. A., & Rivera-Rodriguez, A. J. (2013). SEIPS 2.0: A human factors framework for studying and improving the work of healthcare professionals and patients. Ergonomics, 56(11), 1669-1686. https://doi.org/10.1080/00140139.2013.838643
Institute for Safe Medication Practices. (2020). Guidelines for optimizing safe implementation and use of smart infusion pumps.
How this NU515 Unit 2 example is structured
Scope comes first and is narrow on purpose: one high-alert task, from order verification to a documented running rate, observed ten times in each condition on one unit. The current-state map precedes the new one so every change can be traced to a specific step. Steps are then sorted by type, care, verification and documentation, which shows that integration removed transcription while keeping the double check hospital policy still requires. The failure path gets equal weight with the success path, because three fallbacks in ten starts are where a nurse's time and a patient's risk move. Each fallback is traced to its cause: wireless dead zones in two rooms and a library entry whose units did not match the order. What the maps cannot show closes the paper: night staffing, multi-channel pumps and titrations.
Get an NU515 Unit 2 example written to your instructions
Pick one task a technology changed on your unit, or soon will, and describe how it runs now. Along with the Unit 2 prompt and rubric, that description becomes a model analysis with both maps, timings and the failure path, back in 24-48h and free the first time. The task you map, and what you see, stay yours. 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.
NU515 Unit 2 questions, answered
Do I need to time real observations?
Where possible, yes, even a handful, but many sections accept estimated times if the basis is stated. The sample reports ten composite observations per condition and labels the sample as small. What matters most is that each step comes from the real workflow rather than the policy manual, since policies describe the path nurses are supposed to take, not the one they do.
Which mapping notation should the analysis use?
Swimlane diagrams work well because they show who or what performs each step, and handoffs between lanes are where delays and errors tend to occur. The sample uses five lanes, treating the pump and the record as actors. Some sections prefer flowcharts or value stream maps. Whichever notation is used, a legend and consistent symbols across both maps make comparison possible.
Why include failed auto-programming attempts?
Because the fallback path is part of the real workflow. When integration fails, the nurse programs manually, often under more time pressure than before, and the analysis would be incomplete without that path. The sample shows three failures in ten starts and traces each to a cause. Leaving them out would overstate the benefit and hide the problem later units try to fix.