Revision notes · Lecture 3 · Free
Literature review
In one line: seek widely, retrieve precisely, appraise critically, and keep searching until the study ends.
What this lecture's questions test
Out of the 30 questions on this lecture:
| Idea | Questions |
|---|---|
| Seeking, retrieval, appraisal and databases | 11 |
| Why, when and how to review, and its ethics | 9 |
| Boolean operators | 5 |
| PubMed, MeSH and other databases | 5 |
1. Four terms this lecture keeps using
| Term | What it is |
|---|---|
| Information seeking | Scanning widely, by hand or by computer, to gather potentially useful articles and books |
| Information retrieval | Narrowing a large collection to the items most relevant to your question. You define the query |
| Critical appraisal | Judging, systematically, whether a study is valid and unbiased |
| Database | An organised, searchable collection of articles, abstracts, proceedings, books and citations |
Remember it: seeking is the wide net. Retrieval is the fine sieve. Appraisal is the taste test.
Trap: "scanning the literature to find potentially useful articles" is seeking, not retrieval, even though retrieval sounds right.
2. Boolean operators
| Operator | Does | Results |
|---|---|---|
| AND | Both terms must appear | Fewer: narrows |
| OR | Either term will do | The most: broadens. Used to join synonyms of one idea |
| NOT | Leaves a term out | Fewer |
- Diabetes in children → diabetes AND children.
- Every word for a child → child OR paediatric OR adolescent.
- Back pain but not sciatica → back pain NOT sciatica.
3. PubMed and MeSH
- PubMed is free. It is run by NCBI at the US National Library of Medicine and searches MEDLINE.
- MeSH = Medical Subject Headings, the Library's controlled vocabulary for indexing MEDLINE and PubMed.
- EMBASE uses its own thesaurus, EMTREE, not MeSH.
- Databases the course names include MEDLINE, EMBASE, CINAHL and Google Scholar.
4. Why, when and how
- Why: to avoid repeating work, understand the subject, and find gaps, questions and methods.
- When: at the start, and throughout the project.
- Steps: search → organise → synthesise → find the gaps (lacunae) → develop the research question.
- Where it appears in a paper: introduction, methods and discussion, not results.
- It does not decide your own study's findings or conclusions.
- Ethics: critique a study's weaknesses in a scholarly way. Never skip them, and never attack the authors' reputation.
Watch out: one scenario question sits between critical appraisal and a neighbouring term. The key and standard teaching disagree; the app explains it on that question.
Last look before the exam
- Seeking = wide. Retrieval = relevant. Appraisal = valid?
- OR gives the most results. AND narrows.
- MeSH → PubMed and MEDLINE. EMTREE → EMBASE.
- PubMed is free.
- Keep searching throughout. Literature goes in the introduction, methods and discussion, not results.