This is the stage the course is named for, and it is where the highest-weighted rows usually sit. A differential is not a list of everything the symptom could represent; it is a ranking, and the ranking has to be argued from what is actually common in an outpatient population, adjusted by the features this patient does and does not have. Alongside the ranking sits a second obligation: the dangerous possibilities have to be named and addressed even when they are unlikely. Your section may print this as NR 576 or NR576; it is the same course. Chamberlain publishes no syllabi outside Canvas. The placement here is our teaching judgment from the course's catalog arc; your section's rubric decides what your week actually asks. Clinical hours, logs and preceptor evaluations remain your own record and are never drafted, reconstructed or estimated with help.
What a ranked differential has to argue
Take twenty student write-ups and audit only the differential sections, and the same defect appears in most of them: three or four diagnoses listed with a sentence of definition each, no order that means anything, and no feature from the actual patient attached to any of them. That is a vocabulary exercise. What the rows are scoring is the movement of probability, which requires a starting point, a set of features, and an explicit statement of which direction each feature pushed.
The starting point in primary care is prevalence in a clinic population, and this is where students trained in acute settings most often go wrong. Hospital and emergency department populations are pre-selected by triage, so the mix of causes behind any given complaint is completely different from the mix walking into an office. A differential led by the conditions that dominate inpatient teaching, applied to an ambulatory presentation, is a reasoning error a grader in this course will identify immediately, and it is the single most consistent distinguishing feature between a note that reads as ambulatory and one that reads as imported.
Ranked does not mean confident. The strongest sections say why the leading candidate leads and also what would displace it, which converts the differential from a static list into something that behaves. Naming the finding that would reorder your ranking is doing two jobs at once: it demonstrates the reasoning and it sets up the testing and safety-net sections that follow.
The serious-but-unlikely category deserves its own paragraph rather than a mention. In an office the dangerous causes of a common complaint are usually rare, and the work is not to chase them but to say explicitly what makes each unlikely today, or what would be required to exclude it, and under what circumstances that threshold changes. A write-up that reaches a confident common diagnosis without ever addressing the serious alternative is weaker than one that names the alternative and argues it down.
The boundary applies as always. The encounter, the hours it counted toward and the evaluation of your performance belong to you and to your preceptor; what is being built here is the written reasoning about a patient you genuinely saw, de-identified before a word of it reaches an academic file. Deliverables at this stage are commonly a case analysis with a differential section, sometimes with a comparison table, and any accompanying post is final copy since posts do not reopen after submission in Canvas.
The NR-576 Week 3 method, step by step
Six moves that produce a ranking rather than a list.
-
Write the problem representation and generate from it, not from the symptom
An older adult with treated hypertension and three weeks of progressive exertional breathlessness generates a different candidate set from breathlessness alone. The context and the tempo are half the constraint.
-
Start the ranking from what is common in this setting
Ask what most often causes this presentation in an ambulatory population of this age band, and say so in the text. That sentence is the base rate your subsequent adjustments operate on, and without it the ranking has no origin.
-
Move each candidate with a named feature
This finding raises it. This absence lowers it. Write the direction explicitly for at least three candidates, because a grader scoring differential reasoning is looking for exactly those sentences and nothing else substitutes.
-
Give the dangerous possibilities their own paragraph
Name them, say what makes each unlikely today, and state what finding would change that judgment. This is also the paragraph that makes your safety net believable later.
-
Name the discriminator you still lack
The one piece of information that would most reorder the ranking. Identifying it is what connects the differential to the testing section, and it is the sentence most write-ups leave out.
-
Commit to a working diagnosis and say how sure you are
A ranked list with no verdict has not finished the task. State the working explanation, the confidence you hold it with, and what would change it.
A layout and word budget for the differential section
Our frame for the differential portion of an ambulatory case document, sized for roughly 500 to 750 words. It is our own outline rather than anything the university issues, and your week's rubric outranks it wherever the two disagree.
| Element | What belongs in it | Word target |
|---|---|---|
| Base rate sentence | What most commonly explains this presentation in an outpatient population of this age band, with a source. | 50 to 70 |
| Leading candidate | The features that raise it, the features that would argue against it, and why it leads on balance. | 130 to 180 |
| Second and third candidates | Each with the specific finding or absence that positions it, not a definition of the condition. | 170 to 230 |
| Serious possibilities | The dangerous causes named, what makes each unlikely today, and the finding that would change that. | 110 to 160 |
| The missing discriminator | The single piece of information that would most reorder the list, and why. | 40 to 60 |
| Working diagnosis and confidence | The commitment, the strength of it, and the condition under which you would abandon it. | 60 to 90 |
Evidence craft for probability arguments
Source your base rates and say which population they describe. Studies reporting the distribution of causes for a given complaint in primary care exist for many common presentations and are exactly the right evidence here. Name the setting the figure came from, since a distribution derived from an emergency department cannot carry an argument about a clinic.
Express prevalence as a fraction of a defined group. Roughly one in ten patients presenting with this complaint in general practice, in the cited study, is usable. Common is not. The habit also protects you in the testing section, where the value of a result depends entirely on where you started.
Prefer sources that quantify how much a feature moves probability. Work on the diagnostic performance of individual findings gives you the language of shift rather than association, and shift is what a ranking argument runs on. Where only associational evidence exists, keep your verb honest and say the feature is associated with rather than raises the likelihood by.
Be explicit about the population you are reasoning in. Age band, sex where relevant, comorbidity burden and setting all change the candidate set. One sentence naming those constraints at the top of the section prevents most of the errors that follow.
Keep the de-identification discipline inside the differential. Rare conditions plus specific ages plus occupations can identify a person even without a name. Where a case is unusual, generalize the details that are not doing diagnostic work and say you have done so.
Five mistakes that cost points at this stage
- A hospital differential in a clinic note. Ranking by severity rather than by frequency is the clearest sign the ambulatory shift has not happened.
- Definitions where discrimination was asked for. A paragraph explaining what a condition is tells the grader nothing about this patient.
- An unordered list. Three possibilities with no argument about which leads has skipped the entire graded task.
- The dangerous cause omitted. Silence about a serious alternative reads as not having considered it, which is worse than considering and dismissing it.
- No commitment at the end. A section that surveys without concluding leaves the plan section with nothing to follow from.
Before you submit
- A base rate sentence appears with a source and a named population
- Every candidate carries at least one patient-specific feature moving it up or down
- The direction of each feature's effect is stated, not implied
- Serious possibilities have their own paragraph with what would change the judgment
- The missing discriminator is named
- A working diagnosis is committed to with a stated level of confidence
- Nothing in the section could identify the patient
Building an NR-576 differential this week?
Send the rubric and your de-identified case notes out of Canvas. A premium original draft comes back in 24 to 48 hours with the ranking argued from outpatient frequency and every candidate moved by a named feature, and revisions run until the grade lands.