Three resident narratives that illustrate the narrowing process
Story 1: Maya’s “population‑level” curiosity
Maya entered her second year eager to explore the rising prevalence of chronic kidney disease (CKD) among older adults. She began by scanning systematic reviews and quickly identified a massive body of literature describing risk factors, outcomes, and health‑system costs. Her initial question—“What are the determinants of CKD progression in patients over 65?”—was undeniably important, yet it spanned dozens of specialties and required data that no single service could provide within six months.
To impose order, Maya listed every variable she found interesting (blood pressure, proteinuria, medication adherence, socioeconomic status) and then tried to map each onto the rotations she would cover in the next half‑year. She realized that she would be spending most of her time on inpatient medicine, with only occasional outpatient clinics.
When she presented this draft to her research mentor, the mentor asked Maya to consider a “slice” of the larger picture that could be observed during her inpatient service. Maya pivoted to a narrower focus: the association between early nephrology consultation and length of stay for patients admitted with acute kidney injury (AKI). This version could be answered with chart review of patients she would encounter daily.
What Maya learned was that a feasible question often lives at the intersection of clinical relevance, data accessibility, and the resident’s immediate environment. By aligning the scope with her scheduled rotations, she turned a sprawling epidemiologic project into a manageable retrospective cohort study.
Story 2: Jamal’s “intervention‑first” mindset
Jamal was fascinated by the potential of point‑of‑care ultrasound (POCUS) to improve diagnostic accuracy for pulmonary embolism (PE). He imagined a trial comparing POCUS‑guided pathways to standard imaging protocols across the entire hospital. After weeks of reading guideline statements, he felt overwhelmed by the need for multiple sites, ethics approvals, and a large sample size.
His first attempt to narrow the topic involved drafting a PICO: Population = all inpatients with suspected PE; Intervention = POCUS; Comparison = CT angiography; Outcome = time to diagnosis. The resulting question was still too broad for his timeline. Jamal then asked himself what piece of the pathway he could influence directly.
He discovered that his night‑float team frequently performed bedside lung ultrasounds but rarely documented a formal “PE rule‑out” algorithm. Jamal reshaped his question to: “Does a standardized three‑view lung ultrasound protocol reduce the need for CT in patients with low‑to‑moderate pre‑test probability for PE on the night float service?” This question required only a prospective audit of his own team’s practice, a brief educational session, and a simple data collection form.
The lesson Jamal took away was that narrowing can be driven by the intervention’s feasibility as much as by the population. By focusing on a specific workflow he could control, he created a project that fit within his clinical duties and still contributed novel insight.
Story 3: Priya’s “gap‑finder” approach
Priya loved reading the latest meta‑analyses on antimicrobial stewardship. She wanted to add to the conversation by investigating the impact of rapid diagnostic testing on vancomycin use in intensive care units (ICU). Her literature search returned dozens of multicenter trials, many of which reported conflicting results.
She tried to locate a “research gap” by noting that most studies examined adult ICUs in tertiary centers, while her community hospital lacked a rapid PCR platform. Priya hypothesized that the absence of such technology might lead to longer vancomycin courses. However, she realized she had no direct access to the rapid test, and purchasing it for a research project was beyond her budget.
Instead of abandoning the idea, Priya reframed her question to explore the existing stewardship process: “What is the average duration of vancomycin therapy for suspected MRSA infections in the ICU before and after implementation of a pharmacist‑driven review protocol?” This pivot leveraged a quality‑improvement initiative already underway, allowing her to use existing pharmacy logs and electronic health record data.
Priya’s key takeaway was that a perceived gap in the literature does not always require new technology; sometimes the gap can be examined through a process change already present in the institution.
Key turning points that turned broad ideas into doable questions
Across the three narratives, several decisive actions repeatedly emerged as catalysts for narrowing. The table below aligns each action with the resident who employed it, illustrating how the same strategy can appear in different clinical contexts.
| Action | Resident who used it | Resulting focus |
|---|---|---|
| Map variables to scheduled rotations | Maya | Inpatient AKI length‑of‑stay study |
| Identify a workflow you can directly modify | Jamal | Night‑float POCUS protocol audit |
| Leverage an existing quality‑improvement program | Priya | Pharmacist‑driven vancomycin review analysis |
Beyond the concrete steps, three conceptual shifts deserve special attention:
- From population breadth to service depth. Instead of asking what affects an entire disease cohort, ask what you can observe within the service you actually rotate through.
- From ideal intervention to realistic implementation. Dream projects are valuable for inspiration, but feasibility hinges on whether the intervention is already part of your daily practice or can be introduced with minimal disruption.
- From literature‑driven gap to process‑driven gap. A missing piece in the published record can sometimes be examined by looking at how your own institution currently operates.
“I thought I needed a brand‑new dataset to be original. The moment I realized I could study the practice already happening on my floor, the project stopped feeling impossible.” – Jamal, second‑year internal medicine resident
How to recognize which narrative mirrors your own situation
Begin by sketching a quick three‑column table on a piece of paper: Column A lists the broad topic you are drawn to; Column B notes the clinical settings you will occupy over the next six months; Column C records the resources (data sources, mentors, institutional initiatives) already at your disposal. Then answer the following prompts, matching your answers to the stories above.
If your column B is dominated by inpatient wards and you have easy access to chart review, you may relate most closely to Maya’s experience. Your next step would be to ask, “What outcome can I measure with the patients I see each day?”
If you spend a substantial portion of your schedule on a specific team that uses a particular diagnostic tool or protocol, Jamal’s approach may resonate. Consider whether you can formalize or audit that tool’s use rather than invent a new one.
If you are aware of an ongoing quality‑improvement or stewardship effort, Priya’s strategy could be a good fit. Look for data that the program already collects and think about how you might analyze it through a research lens.
Identifying the closest match helps you avoid reinventing steps you have already taken and directs you toward the most efficient path forward.
Practical actions you can start this week
Even with a heavy rotation schedule, a handful of focused activities can move a vague idea toward a concrete question.
- Schedule a 15‑minute “scope‑check” with a faculty mentor. Bring a one‑page summary that includes your broad interest, the patient population you will encounter, and any existing data streams you can tap.
- Conduct a “data inventory” during a single shift. Note every chart element, lab result, or procedural note that is routinely recorded and could serve as a variable.
- Write a provisional PICO statement, then immediately ask: “Which component of this PICO aligns with a service I will be on for at least two weeks?” If the answer is “none,” rewrite the statement until the answer is affirmative.
- Locate a recent departmental newsletter or quality‑improvement dashboard. Highlight any metric that mentions your area of interest; this often points to a ready‑made dataset.
- Draft a brief timeline that pairs each project milestone (IRB draft, data extraction, analysis) with a specific rotation block. Visualizing the overlap reveals whether the plan is realistic.
These steps are deliberately brief; the goal is to generate a question that can survive the inevitable interruptions of clinical duties.
Adapting the approach to your unique rotation schedule
Every resident’s calendar contains a mixture of predictable blocks (e.g., two weeks of general medicine) and variable components (e.g., on‑call nights). To keep the research question flexible, embed “contingency clauses” into your design. For instance, if you plan a retrospective chart review, specify a primary data source (your inpatient service) and a backup source (the hospital’s electronic health record query tool) in case a rotation change limits access.
When you anticipate a future rotation that might provide richer data—such as an outpatient clinic—frame your question with an “extension” option. An example could be: “Primary analysis will use inpatient data; a secondary analysis will explore outpatient follow‑up if time permits.” This language signals to mentors and IRBs that the project is scalable without committing you to an unattainable scope.
Finally, remember that the scholarly project deadline is a hard boundary, but the learning process is not. If, after a month, the chosen question still feels too expansive, revisit the three narratives and the key turning points. Adjusting the focus a second time is common and does not reflect failure; it reflects the iterative nature of clinical research.
By tracing the arcs of Maya, Jamal, and Priya, you now have a menu of strategies that align research ambition with the practical realities of residency. Choose the narrative that feels most familiar, apply the corresponding turning points, and use the weekly actions to solidify a question that fits within your six‑month window. With a question in hand, the subsequent steps of protocol writing, data collection, and analysis become a series of manageable tasks rather than an overwhelming mountain.