NR-730 Week 4 The Search Trail and Evidence Table: How to Write It

What NR-730 Week 4 asks for

A doctoral student building a sepsis recognition project in an emergency department spent three weeks reading and arrived with 41 saved articles and no way to see them. She could describe any one of them from memory. She could not tell you how many were conducted in emergency settings, how many measured time to antibiotic administration rather than mortality, or which two contradicted each other. The reading was real. The structure was missing, and without structure a synthesis becomes a sequence of paragraphs each summarizing one study, which is the most common weak shape in doctoral project writing.

A documented search fixes the first half of that. Write it as a method: which databases, which terms and combinations, which filters, which date range, what was returned, what was screened out at title and abstract, what was read in full, and what remained. Report the numbers. A search that yielded 612 records, screened to 58 full texts and retained 19, is a transparent process; a search described as a thorough review of the literature is an assertion. Doctoral readers score this row on specificity because reproducibility is the difference between a search and a browse.

The table fixes the second half. Columns should be chosen so that reading down them answers the questions your synthesis needs: citation, design, setting, sample with its size, the intervention as actually delivered, comparator, outcomes measured, key results with numbers, quality appraisal, and relevance to your setting. That last column is the one students omit and the one navigators ask for, because a project's evidence table has a job that a research paper's does not, which is establishing that this change belongs in this environment.

One more thing shapes this stage. Translation work favors evidence that is already consolidated. Guidelines, systematic reviews and syntheses do more for a DNP project than an assortment of primary studies, because they represent the settled position you are moving into practice. Build the table around the strongest consolidated evidence first, then use primary studies and implementation reports to supply the practical detail: what was actually delivered, by whom, with what training and at what cost.

The NR-730 Week 4 method, step by step

Seven moves for a search you can defend and a table you can use.

1. Read the rubric for whether the search method is separately scored

   It usually is, and it is the easiest row in the course to score well because it rewards documentation rather than insight. Find out whether a diagram or a count of records is expected before you start searching, since reconstructing counts afterward is painful.

2. Build the search from your question's parts

   Population, intervention, comparison and outcome each supply terms. Write the concept groups first, then the synonyms inside each group, then the combinations. A search built by typing a sentence into one box will miss whole bodies of work with different vocabulary.

3. Search at least two databases and say which

   Nursing and medical databases index differently, and a guideline repository will surface consolidated sources neither covers well. Name each database, the interface you used and the date you ran it, because indexing changes over time and a dated search is a defensible one.

4. Record counts at every stage as you go

   Returned, screened by title and abstract, retrieved in full, excluded with reasons, retained. Keep a running note. This is fifteen minutes of bookkeeping during the search and two evenings of reconstruction afterward if you skip it.

5. Set your table columns before you extract anything

   Decide what the synthesis will need to say and build the columns to serve it. Adding a column halfway through means revisiting every article already extracted, which is where most of the wasted time in this stage goes.

6. Extract numbers, not impressions

   Sample sizes with their setting, effects in real units, follow-up windows, attrition. A results cell that says the intervention was effective has thrown away the information that would let you compare studies to each other.

7. Fill the relevance column honestly

   Say what differs between each study's setting and yours: academic versus community, physician-led versus nurse-led, a unit with an intensivist on site versus one without. Those differences are what your later implementation reasoning will be built from.

A layout and word budget for the search and table section

Our frame for a documented search with an evidence table, sized for roughly 900 to 1,200 words of prose beside the table itself. It is our own outline rather than anything the university issues, and your week's rubric outranks it wherever the two disagree.

Evidence craft for search documentation

Give the search a date and keep it current. Searches age. Record the date each database was run and, if months pass before your final document, re-run the strategy and report the update rather than presenting a stale set as current.

Use a named appraisal tool and say which. Structured critique instruments exist for different designs, and applying a named one with its year makes your quality column checkable. Rating studies as good or fair without an instrument is a personal judgment wearing an evaluative label.

Report exclusions with reasons, in counts. Thirty-nine excluded at full text because the population was pediatric, eleven because the outcome was not measured, six because no comparison group existed. Reasons in counts are what let a reader see whether the retained set was shaped by the question or by convenience.

Treat improvement reports as their own category. Single-site quality improvement reports carry practical detail that trials omit, and they cannot rule out what else changed at the same time. Label them clearly in the design column and use them for feasibility rather than for effect size.

Do not let the table quietly become a summary. If cells are filling with sentences, the columns are wrong. Extract into short factual entries so that reading down a column produces a comparison rather than a reread.

Five mistakes that cost points in this week's territory

• An undocumented search. No databases, no terms, no counts, no date, which leaves the entire evidence base unreproducible and the method row unscoreable.
• One database only. A single index will miss work using different vocabulary, and the omission is visible to any reader who knows the field.
• A table of narratives. Cells full of prose defeat the purpose, because the point of the table is that columns can be compared.
• No appraisal instrument. Quality judgments without a named tool cannot be checked and read as impressions.
• No relevance column. Without it, the synthesis has no basis for arguing that evidence from other settings should be trusted in yours.

Before you submit

• Every database, term combination, filter and search date is recorded
• Counts appear at each screening stage with reasons for exclusion
• The table's columns include setting, sample size, intervention as delivered and results in real units
• A named appraisal instrument is applied and cited
• A relevance column names what differs between each study's setting and yours
• The limits of the search, including what was not accessible, are stated

Questions students ask about this stage

Keep going