| Course | NR 512 Fundamentals of Nursing Informatics |
|---|---|
| Week | 5 |
| Paper type | Information model write-up |
| Length | About 1,170 words, 4 pages |
| Format | APA 7 student paper |
| School | Chamberlain University |
| Program | MSN, Nursing Informatics |
| Updated | September 2026 |
Free sample paper for NR 512 Week 5
Information Model Write-Up: Five Entities Behind Oxygen Therapy and Pulse Oximetry Charting on a Medical Unit, and What the Shape Makes Easy and Impossible to Ask
Student Name
College of Nursing, Chamberlain University
NR 512: Fundamentals of Nursing Informatics
Instructor Name
Month Day, Year
Information Model Write-Up: Five Entities Behind Oxygen Therapy and Pulse Oximetry Charting on a Medical Unit, and What the Shape Makes Easy and Impossible to Ask
Scope
The model covers one area of nursing documentation on an adult medical unit: supplemental oxygen and the pulse oximetry readings taken while it is in use or after it stops. It does not cover the oxygen order's approval, respiratory therapy billing, ventilators or blood gases. Keeping the scope this small is deliberate. The questions this model has to answer are the ones a nurse, a charge nurse and a quality analyst ask about oxygen and saturation, and a larger model would bury those decisions under entities nobody on the unit uses.
The approach follows the entity-relationship method described by Chen (1976): things the record needs to hold facts about become entities, the facts become attributes, and the links between them become relationships with a stated cardinality in both directions.
The Entities and Their Attributes
Table 1 lists the five entities, their attributes and data types. Units of measure have their own attribute wherever a number is stored.
Table 1
Entities, Attributes and Data Types
| Entity | Key attributes (data type) |
|---|---|
| Encounter | encounter ID (identifier); patient ID (identifier); admit time (date-time); unit (code) |
| Oxygen episode | episode ID (identifier); encounter ID; device type (code: nasal cannula, simple mask, Venturi, high-flow, blow-by, none); start time (date-time); end time (date-time, empty while running) |
| Device setting | setting ID; episode ID; flow (decimal); flow unit (L/min); inspired oxygen concentration (decimal, %); effective time (date-time); entered by (staff ID) |
| Saturation observation | observation ID; encounter ID; value (integer); unit (%); effective time (date-time); entry time (date-time); method (code: spot check, continuous); probe site (code); status (preliminary, validated, entered in error); validated by (staff ID) |
| Oxygen context | observation ID; flow at reading (decimal, L/min, 0 for room air); concentration at reading (decimal, %); device at reading (code) |
Relationships and Cardinality
Encounter to oxygen episode. One encounter has zero or many oxygen episodes; each episode belongs to exactly one encounter. Zero matters: many patients never need oxygen, and the model must not force an empty episode on them.
Oxygen episode to device setting. One episode has one or many settings; each setting belongs to exactly one episode. Flow is turned up, down and back again within a single episode, so the settings cannot be attributes of the episode without losing every change but the last.
Encounter to saturation observation. One encounter has zero or many observations; each observation belongs to one encounter.
Saturation observation to oxygen context. One observation has exactly one oxygen context, and each context belongs to exactly one observation. This is the relationship that matters most, and it is stated in both directions because the reverse is where designs usually break: a context without its observation is meaningless, and an observation without its context is ambiguous.
The Decisions Behind the Shape
Why device setting is its own entity. A first draft put flow on the oxygen episode. It failed as soon as a realistic night was walked through: a patient started at 2 L/min at 2200, went to 4 L/min at 0200 and back to 3 L/min at 0600. With flow on the episode, only 3 L/min would survive, and the 0200 increase, the clinically important event, would vanish.
Why oxygen context is stored on the observation instead of looked up. In principle, the flow at the time of any reading could be found by matching the observation's effective time to the device setting in force at that moment. That design stores nothing twice, but it fails in practice in two ways. Settings are often entered a few minutes after they are changed, so a lookup by time can return the old flow. And if a setting is later corrected, the lookup silently changes the context of readings that were interpreted with the old one. Storing a copy of the context with each observation, confirmed by the nurse at the time of saving, accepts some duplication in exchange for a record of what was known when the reading was used. It also matches the HL7 pulse oximetry profile, which carries flow rate and concentration as components of the saturation observation itself (HL7 International, n.d.).
Why room air is a value, not an empty field. The flow attribute of the oxygen context holds 0 for room air. An empty field is then reserved for one meaning only, not recorded, which the model can detect and a report can count. The HL7 profile recommends the same convention (HL7 International, n.d.).
Why two times. Effective time says when the reading happened; entry time says when it was charted. Without both, a reading charted an hour late would appear in the record as though it had been known an hour earlier.
Why status. Monitor-integrated readings arrive preliminary and become validated only when a nurse accepts them. A query that mixes the two would count artifact. On a unit with continuous monitoring, preliminary readings outnumber validated ones many times over, and most of them are never looked at by anyone. Keeping status as an attribute lets a report choose which kind it counts, and lets an audit ask how many preliminary readings below 90% were never reviewed, a question that would be unanswerable if unvalidated values were discarded or silently promoted.
What the Model Makes Cheap
Several questions become one simple query each: every saturation below 92% recorded on room air in the past 24 hours; the highest flow any patient on the unit needed overnight; how long each oxygen episode ran; whether a patient's saturation was checked after each increase in flow, and how soon. The last question is useful for early warning scoring, since NEWS2 treats being on oxygen as a scored parameter in its own right (Royal College of Physicians, 2017), and a unit can check whether every scored reading had its oxygen status recorded.
What the Model Makes Impossible
The model cannot say why a nurse chose one monitor value over another in a given hour, because it stores only the validated value and not the rejected candidates. It cannot say whether the patient was asleep, eating or moving at the time of the reading. It cannot distinguish oxygen that was flowing from oxygen that reached the patient, which is why the data definition for flow must say which one to record. And it holds no information about skin pigmentation, perfusion or other factors known to affect oximeter accuracy. Each of these could be added as an attribute, and each addition costs charting time. The model leaves them out on purpose and says so, so that nobody later assumes the data can answer questions it was never built to answer.
References
Chen, P. P.-S. (1976). The entity-relationship model: Toward a unified view of data. ACM Transactions on Database Systems, 1(1), 9-36. https://doi.org/10.1145/320434.320440
HL7 International. (n.d.). US Core pulse oximetry profile (US Core Implementation Guide, Version 9.0.0, STU 9). Retrieved September 30, 2026, from https://hl7.org/fhir/us/core/StructureDefinition-us-core-pulse-oximetry.html
Royal College of Physicians. (2017). National Early Warning Score (NEWS) 2: Standardising the assessment of acute-illness severity in the NHS. Updated report of a working party. https://www.rcp.ac.uk/improving-care/resources/national-early-warning-score-news-2/
What the NR 512 Week 5 assignment asks for
The Week 5 assignment asks you to model the data behind a small area of nursing documentation. Typical prompts ask you to identify entities, attributes and relationships, show them in a diagram such as an entity relationship diagram, and explain your choices. Some sections ask how the model supports data retrieval, reporting or decision support. Keep the scope small: a single assessment, one order and its results, or one kind of therapy. The course expects you to understand the difference between a screen and the data under it. Check whether your section wants a particular notation, and whether the diagram can be submitted as a table or must be an image.
How this NR 512 Week 5 example is built
The example limits itself to supplemental oxygen and the saturation readings taken while it runs. It uses the entity relationship approach described by Chen, and Table 1 lists five entities with their attributes and data types. Relationships are stated both ways, including zero cases, such as an encounter with no oxygen episodes. The decisions section works through a realistic night to show why flow cannot be an attribute of the episode, weighs storing oxygen context against looking it up, and explains why room air is a value, why two timestamps are needed and why each reading carries a status. The paper then lists questions the model answers in one query and the questions it cannot answer at all.
NR 512 Week 5 rubric: what full points look like
Rubrics for the information model usually give most of their points to correctness of the entities, attributes and relationships, and to the explanation of design choices. Cardinality stated in only one direction is a common reason for lost points. Graders look for data types and units, since claims about what a model can answer depend on them, and for timestamps wherever order matters. A criterion on application asks how the model supports practice or reporting, and the sample answers it with a list of queries. Some sections also grade the diagram's notation. APA 7 format and clear writing complete the rubric, with citations for the modeling method and any standard used.
NR 512 Week 5 help from the desk
The most costly mistake is modeling the screen instead of the data. A form is a view, and copying its layout produces duplicated data and missing relationships. Another is stating cardinality in one direction, which hides the case where the design breaks. Leaving out timestamps makes repeated observations impossible to order. Students also let the scope grow, often by adding billing or staffing entities, which turns a focused model into a sketch of the whole hospital. Name each attribute's data type and unit. Finally, say what your model cannot do. Graders read that section as evidence that you understand every design rules some questions out, and it is often the difference between a good paper and a strong one.
Get a NR 512 Week 5 example written to your instructions
Send the Week 5 prompt, the area of documentation your section assigned, and the rubric. We write the model and the diagram to your instructions, keep the scope small enough to argue properly, and include the answerable and unanswerable questions at the end. It comes back inside 24-48h, and the first one is free.
More NR 512 papers and MSN samples
- NR 512 Week 1: Discussion Post on One Recorded Item
- NR 512 Week 2: Data Definition Brief, Oxygen Flow
- NR 512 Week 3: Terminology Comparison for One Finding
- NR 512 Week 4: Concept Map of One Reading's Path
- NR 512 Week 6: Standards Appraisal
- NR 512 Week 7: Documentation Case Analysis
- NR 512 Week 8: Informatics Position Paper
NR 512 Week 5 questions, answered
How formal does the notation need to be?
Most sections accept a clean box and line diagram without requiring a particular notation, though consistency inside the write-up is expected. If a formal notation is named in the instructions, use it exactly and say which one in the legend. Where none is named, define your symbols in a short legend, since a line whose meaning has to be guessed costs the diagram its evidence.
How small should the modeled area be?
Smaller than instinct suggests. Four to six entities is enough to show relationships, cardinality and a genuine design decision, and it leaves room to write about those decisions instead of listing boxes. Large models eat the word count with names alone and produce write-ups that describe a system without arguing anything about meaning.
Can I model something from my workplace?
Yes, with the care the rest of the course asks for. Model the structure, not the institution: no facility name, no vendor claim, no real values. An employer's configuration belongs to the employer, and the parts worth writing about are the generic ones anyway, because the design questions are the same wherever that form happens to live.
Where can I find a free NR 512 Week 5 sample paper?
You can read a complete NR 512 Week 5 information model write-up on this page, with its entity table and design decisions. If you are modeling a different area, describe it on the request page and a matching write-up is drafted for you at no cost.
Can the NR 512 information model be shown as a table?
Many sections accept a table of entities and attributes alongside a diagram, and some accept a table alone. The sample uses a table for entities and states the relationships in prose. Check your section's instructions before choosing.