lancet-research-in-context
GitHub为柳叶刀原创研究文章构建强制性的Research in Context面板，包含文献证据、新增价值及综合影响三部分。基于系统检索结果，确保内容规范且符合期刊要求。
触发场景
安装
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill lancet-research-in-context -g -y
SKILL.md
Frontmatter
{
"name": "lancet-research-in-context",
"description": "Use to build The Lancet's signature Research in context panel — the three mandatory headed parts (Evidence before this study, Added value of this study, Implications of all the available evidence), grounded in a systematic literature search with stated databases, terms, and dates. Mandatory for original research Articles."
}
Research in Context Panel (lancet-research-in-context)
When to trigger
- Writing an original research Article for The Lancet (the panel is mandatory).
- There is no panel, or it reads like a discussion paragraph rather than the three required parts.
- The "Evidence before this study" part asserts what was known without a documented systematic search.
- New evidence appeared during review and the panel needs updating.
Why this is the most distinctive Lancet skill
The Research in context panel is The Lancet's signature artifact and is required for every original research Article. It forces authors to place their study inside a systematically searched evidence base — not a cherry-picked introduction. Editors and reviewers read it early; a weak or search-free panel signals an unsystematic literature review.
The panel has EXACTLY three headed parts
1. Evidence before this study
What was known before you did this study, and how you found it. This must describe a systematic search, not a casual reading:
- The databases searched (e.g., MEDLINE/PubMed, Embase, the Cochrane Library, regional databases such as LILACS/Global Index Medicus for global-health work).
- The exact search terms / strategy used.
- The date range (and the date the search was last run).
- Any language restrictions and inclusion criteria.
- A brief synthesis of what that evidence showed and what remained uncertain — including the quality/limitations of prior evidence (a GRADE-style judgement is welcome).
2. Added value of this study
What your study adds that the prior evidence did not provide — the specific new knowledge, in one short paragraph. Concrete and non-promotional: the new effect estimate, the new population/setting, the resolved uncertainty.
3. Implications of all the available evidence
What your findings plus the existing evidence mean together — for clinical practice, public-health policy, and future research. Calibrate causal/practice-changing language to the design. This is where the global-health/equity and policy implications belong.
Exact template
Research in context
Evidence before this study
We searched [databases] for [study types] published between [start date] and
[end date / "the search was last updated on <date>"] using the terms
"[term 1]" AND "[term 2]" [AND "[term 3]"], with [language restrictions / none].
[Inclusion criteria.] [What the existing evidence showed and its key
uncertainties/limitations, ideally with a quality judgement.]
Added value of this study
[In one short paragraph: the specific new knowledge this study provides —
the new estimate, population, setting, or resolved uncertainty.]
Implications of all the available evidence
[What the totality of evidence now implies for clinical practice, for
public-health policy, and for future research — with calibrated causal
language and any equity/global-health implications.]
Quality checks
- All three headed parts present, in order, with the exact headings.
- "Evidence before this study" documents a real systematic search: databases + terms + dates + language.
- The search supports the introduction and the abstract's Background (consistent claims).
- "Added value" is specific and non-promotional.
- "Implications" speaks to practice, policy, and research — not just "more research is needed."
- Total panel is concise (typically a few hundred words; confirm the current word guidance).
Output format
【Panel parts present】 Evidence before / Added value / Implications — all three? yes/no
【Systematic search documented】 databases + terms + dates + language? yes/no (list gaps)
【Consistency】 search matches Introduction + abstract Background? yes/no
【Added value specific?】 yes/no
【Implications cover practice + policy + research?】 yes/no
【Drafted panel】 the three-part panel, filled
【Next】 lancet-abstract
Anti-patterns
- Do not write "Evidence before this study" from memory — it must reflect a documented search with terms and dates.
- Do not merge the three parts into one paragraph or drop a heading.
- Do not overstate "Added value" beyond what the data support.
- Do not end "Implications" on a generic "further studies are warranted" — name the practice/policy consequence.
版本历史
- 1839142 当前 2026-07-05 14:01
