A comprehensive review is only as defensible as its search, and this stage asks you to build one and then write it down so precisely that a stranger could rerun it and land on your numbers. That means a block structure built on controlled vocabulary and free text, a strategy adapted correctly for each database rather than pasted between them, deliberate pursuit of studies that indexes miss, and an accounting of records from first retrieval through deduplication. The register is procedural: exact strings, exact counts, exact dates. Your section may print this as NR 714 or NR714; 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.
What NR-714 Week 2 asks for
An infection preventionist across a group of nursing homes is reviewing antibiotic stewardship interventions in long-term care. Her first search returns two thousand records, nearly all of them from hospitals, because the vocabulary she used belongs to acute care. Her second attempt, restricted with a setting term she invented, returns nine. Neither search is wrong in effort and both are wrong in construction, because she is searching with words rather than with a strategy. Long-term care is indexed under several different headings depending on the database and the decade, and a review that does not account for that vocabulary problem will systematically miss its own literature.
The block method is the backbone. One block per concept in your question, each block containing the controlled vocabulary terms the database provides plus the free text synonyms authors actually use, combined inside the block with OR. Blocks are then combined with AND. Truncation captures word endings, adjacency operators keep phrases together, and field tags decide whether you are searching titles and abstracts or the whole record. The reason this structure matters is recall: a review is expected to find nearly everything eligible, which means accepting a large number of irrelevant records rather than filtering aggressively at the search stage.
Adaptation between databases is where most student searches quietly fail. Controlled vocabulary is database-specific, and a subject heading in one index does not exist in another. Truncation symbols and adjacency syntax differ. A strategy pasted unchanged into a second database returns results, which is precisely the danger, because it looks successful while running with several terms silently disabled. Report each database's strategy in full in an appendix rather than presenting one strategy and claiming it was used everywhere.
Supplementary searching completes the picture. Reference lists of included studies and of relevant reviews, forward citation checking, trial registries, professional and government sources, and dissertation or report literature all catch material the main databases miss, and for post-acute and transitional care topics a substantial part of the useful evidence sits in government and association reports rather than in journals. Expect a written search strategy with the full strings, a record of counts by source, and often a discussion post comparing approaches. Post the actual string; posts do not reopen once submitted in Canvas.
The NR-714 Week 2 method, step by step
Six moves from a protocol question to a search anyone could rerun.
-
Extraction of concepts from the protocol question
Take each element of the structured question and treat it as a search block. Most reviews need only two or three blocks; adding a block for every element usually over-constrains the search and starves the yield.
-
Harvesting of vocabulary from known relevant papers
Find three or four papers you know belong in the review and read their indexing terms and their author keywords. This is the fastest route to the controlled vocabulary and the synonyms your own wording would never have produced.
-
Assembly of each block with headings and free text
Combine subject headings with title and abstract terms inside the block using OR, apply truncation deliberately, and check what each variant retrieves before committing. Over-truncation is a common and invisible source of noise.
-
Translation of the strategy for each database
Rebuild the subject heading layer using each database's own thesaurus and adjust the syntax to its rules. Record the finished strategy for every database separately, with the date it was run.
-
Pursuit of the literature the indexes miss
Check reference lists, run forward citation searches on key papers, look at trial registries, and search the government and professional sources relevant to your setting. Record what each supplementary route contributed.
-
Accounting of the retrieval and deduplication
Report the number of records from each source, the total before deduplication, the number of duplicates removed and the number entering screening. These counts are the opening row of the flow diagram your review will need.
A layout and word budget for a search strategy report
Our frame for the search deliverable, sized for roughly 1,400 to 1,800 words plus the full strategies and count table in an appendix. It is our own outline rather than anything the university issues, and your week's rubric outranks it wherever the two disagree.
| Section | What belongs in it | Word target |
|---|---|---|
| Search objectives | What the search must retrieve, restated from the protocol, and the recall against precision balance you chose. | 160 to 210 |
| Sources and their roles | Each database named with what it uniquely contributes, plus the supplementary routes planned. | 220 to 280 |
| Concept blocks and vocabulary | The blocks, the controlled vocabulary in each, the free text synonyms and where the vocabulary came from. | 280 to 340 |
| Strategy construction | Boolean structure, truncation and adjacency decisions, field tags, limits applied and the reason for each limit. | 260 to 320 |
| Adaptation across databases | What changed for each platform and where an exact equivalent term did not exist. | 200 to 260 |
| Yield and accounting | Counts by source and date run, duplicates removed, records entering screening, and what the yield implies. | 240 to 300 |
Evidence craft for search reporting
Paste the strategy, do not describe it. A line-numbered strategy showing each block and the combination steps is the standard, and prose describing which terms you used is not an acceptable substitute at this level. Put the full strategies in an appendix and reference them from the text.
Record the date each search was run. Databases grow continuously, so a search without a run date cannot be reproduced or updated. If you rerun anything before submitting, report the later date and the revised counts.
Justify every limit you apply. Language, date, publication type and age group limits all remove records, and each one is a decision that could bias the review. Give a reason, and say what the limit might have excluded.
Treat grey literature as a named source, not an afterthought. Government reports, association guidance, quality collaborative publications and registry entries carry real evidence in post-acute topics. Name the specific sources you searched and how, since grey literature searching is frequently reported too vaguely to be reproduced.
Report the uncomfortable numbers. A search that returned four thousand records to yield eleven studies is normal in this work. Reporting the funnel honestly demonstrates comprehensiveness; a suspiciously tidy retrieval suggests a search that was filtered before it was recorded.
Five mistakes that cost points in this week's territory
- One strategy pasted into every database. Subject headings are platform-specific, and a pasted strategy runs with parts of itself silently disabled.
- Keywords only, no controlled vocabulary. Free text alone misses records that are indexed but worded differently, and the omission is obvious to anyone who has searched systematically.
- Filtering for relevance during the search. Narrowing until the result list looks manageable converts a search into a selection and undermines the review's claim to comprehensiveness.
- Counts missing or inconsistent. Numbers that do not reconcile between the text, the table and the flow account are the fastest way to lose a reader's trust in the whole review.
- Grey literature claimed but not specified. Saying grey literature was searched, without naming sources, dates and terms, is an unverifiable assertion.
Before you submit
- Full line-numbered strategies appear for every database searched
- Each strategy uses that platform's own controlled vocabulary
- Every limit applied carries a stated justification
- Supplementary sources are named with what each contributed
- Run dates appear for each search
- Record counts reconcile across text, table and flow account
- Every reference appears in the text and every in-text citation appears in the list
Documenting the NR-714 search this week?
Send the rubric and your protocol out of Canvas. A premium original draft comes back in 24 to 48 hours with block-built strategies, per-database adaptation and a reconciled retrieval account, and revisions run until the grade lands.