literature-review

name	literature-review
description	Conduct systematic literature reviews for research. Use when: writing grant background sections, synthesizing research findings, performing systematic reviews, identifying research gaps, building evidence for proposals, creating annotated bibliographies.

Literature Review

Overview

Conduct systematic, comprehensive literature reviews following rigorous academic methodology. Search multiple databases, synthesize findings thematically, verify citations, and generate professional output documents for grant proposals and research papers.

When to Use

• Writing the background section of grant proposals
• Conducting systematic reviews for publication
• Synthesizing current knowledge on a topic
• Identifying research gaps for proposals
• Building evidence base for significance statements
• Creating annotated bibliographies
• Performing meta-analyses or scoping reviews

Core Workflow

Phase 1: Planning and Scoping

Define Research Question (PICO Framework)

## Research Question

**Population**: Who/what is being studied?
**Intervention**: What treatment/exposure/approach?
**Comparison**: Against what alternative?
**Outcome**: What results are measured?

Example:
"What is the efficacy of machine learning approaches (I) 
for predicting grant success (O) in NIH R01 applications (P) 
compared to traditional review metrics (C)?"

Set Search Parameters

## Search Parameters

**Date Range**: 2015-2024
**Languages**: English
**Publication Types**: 
- Peer-reviewed articles
- Preprints (arXiv, bioRxiv)
- Systematic reviews
**Exclusions**:
- Editorials, letters, opinion pieces
- Non-English publications
- Abstracts only

Phase 2: Literature Search

Database Selection by Domain

Domain	Primary Databases
Biomedical	PubMed, PMC, Embase
General Science	Web of Science, Scopus
Preprints	arXiv, bioRxiv, medRxiv
AI/ML	Semantic Scholar, ACM DL
Social Science	JSTOR, SSRN
Multidisciplinary	Google Scholar

Search Strategy Template

## Search Strategy

### Database: PubMed
**Date Searched**: [Date]
**Search String**:

("machine learning" OR "artificial intelligence" OR "deep learning") AND ("grant" OR "funding" OR "proposal") AND ("prediction" OR "success" OR "outcome") AND 2015:2024[pdat]

**Results**: [N] articles
**Filters Applied**: English, Journal Article

### Database: [Second database]
[Repeat structure]

Search Commands

# PubMed search via Entrez
from Bio import Entrez

Entrez.email = "[e-mail]"

def search_pubmed(query, max_results=100):
    """Search PubMed and return PMIDs."""
    handle = Entrez.esearch(
        db="pubmed",
        term=query,
        retmax=max_results,
        sort="relevance"
    )
    results = Entrez.read(handle)
    return results["IdList"]

# Semantic Scholar API
import requests

def search_semantic_scholar(query, limit=100):
    """Search Semantic Scholar."""
    url = "https://api.semanticscholar.org/graph/v1/paper/search"
    params = {
        "query": query,
        "limit": limit,
        "fields": "title,authors,year,abstract,citationCount"
    }
    response = requests.get(url, params=params)
    return response.json()

Phase 3: Screening and Selection

PRISMA Flow Diagram

┌─────────────────────────────────────────────┐
│     Records identified from databases       │
│                  (n = XXX)                  │
├─────────────────────────────────────────────┤
│ Database 1: n =                             │
│ Database 2: n =                             │
│ Database 3: n =                             │
└────────────────────┬────────────────────────┘
                     │
                     ▼
┌─────────────────────────────────────────────┐
│    Records after duplicates removed         │
│                  (n = XXX)                  │
└────────────────────┬────────────────────────┘
                     │
                     ▼
┌─────────────────────────────────────────────┐
│         Records screened (title)            │
│                  (n = XXX)                  │
├─────────────────────────────────────────────┤
│    Records excluded: n =                    │
│    (Not relevant to topic)                  │
└────────────────────┬────────────────────────┘
                     │
                     ▼
┌─────────────────────────────────────────────┐
│        Records screened (abstract)          │
│                  (n = XXX)                  │
├─────────────────────────────────────────────┤
│    Records excluded: n =                    │
│    (Did not meet inclusion criteria)        │
└────────────────────┬────────────────────────┘
                     │
                     ▼
┌─────────────────────────────────────────────┐
│       Full-text articles assessed           │
│                  (n = XXX)                  │
├─────────────────────────────────────────────┤
│    Full-text excluded: n =                  │
│    Reasons:                                 │
│    - Wrong population: n =                  │
│    - Wrong outcome: n =                     │
│    - No data available: n =                 │
└────────────────────┬────────────────────────┘
                     │
                     ▼
┌─────────────────────────────────────────────┐
│        Studies included in review           │
│                  (n = XXX)                  │
└─────────────────────────────────────────────┘

Phase 4: Data Extraction

Extraction Template

## Study: [First Author et al., Year]

**Citation**: [Full citation]
**DOI**: [DOI]
**Study Design**: [RCT/Cohort/Cross-sectional/etc.]

### Population
- Sample size: n =
- Characteristics:

### Methods
- Approach:
- Key variables:

### Key Findings
- Primary outcome:
- Secondary outcomes:
- Effect sizes:

### Quality Assessment
- Risk of bias: [Low/Medium/High]
- Limitations noted:

### Relevance to Our Research
- [How this informs our work]

Phase 5: Thematic Synthesis

Synthesis Structure (NOT study-by-study)

## Results

### Theme 1: [Major Finding Area]

Multiple studies have examined [topic], with converging evidence 
suggesting [synthesis]. Author et al. (2020) demonstrated [finding], 
which was corroborated by [Author 2021; Author 2022]. However, 
[contrasting evidence] was reported by [Author 2023], who found 
[different result] in [different population/context].

The methodological approaches varied considerably, with [N] studies 
using [method A] and [N] studies employing [method B]. Despite these 
differences, the overall pattern suggests [synthetic conclusion].

### Theme 2: [Second Finding Area]
[Continue thematic synthesis...]

### Theme 3: [Third Finding Area]
[Continue thematic synthesis...]

Phase 6: Gap Analysis

## Research Gaps

### Methodological Gaps
1. Limited use of [methodology] in this domain
2. Insufficient sample sizes in [context]
3. Lack of longitudinal studies

### Knowledge Gaps
1. Unknown whether [relationship] holds in [context]
2. Mechanisms underlying [phenomenon] not established
3. No studies examining [specific population]

### Translational Gaps
1. Findings not validated in real-world settings
2. Implementation barriers not addressed
3. Cost-effectiveness unknown

## How This Proposal Addresses These Gaps
[Connect to your proposed research]

Citation Management

Citation Format Templates

APA 7th Edition:

Author, A. A., & Author, B. B. (Year). Title of article. 
Journal Name, Volume(Issue), Page–Page. https://doi.org/xxxxx

Nature Style:

Author, A.A. & Author, B.B. Title of article. Journal Name 
Volume, Page–Page (Year).

Vancouver (NIH preferred):

Author AA, Author BB. Title of article. Journal Name. 
Year;Volume(Issue):Page-Page. doi: xxxxx

Citation Verification

def verify_doi(doi):
    """Verify DOI resolves correctly."""
    import requests
    url = f"https://doi.org/{doi}"
    response = requests.head(url, allow_redirects=True)
    return response.status_code == 200

def format_citation(metadata, style="apa"):
    """Format citation from metadata."""
    authors = ", ".join(metadata["authors"])
    year = metadata["year"]
    title = metadata["title"]
    journal = metadata["journal"]
    # Format according to style
    pass

Grant Proposal Integration

Background Section Template

## Background and Significance

### Current State of Knowledge
[Synthesize literature establishing the context]

### Critical Gap
"Despite significant progress in [area], a critical gap remains 
in our understanding of [specific gap]. Specifically, [cite 
supporting evidence for the gap]."

### How This Proposal Fills the Gap
"The proposed research will address this gap by [approach], 
building on preliminary findings that [evidence]."

### Impact
"Successful completion of this work will [outcomes], with 
implications for [broader impact]."

Anti-patterns

Mistake	Solution
Study-by-study summaries	Synthesize thematically
Uncritical acceptance	Assess quality and bias
Missing recent literature	Search within last 6 months
Citation errors	Verify all DOIs
Bias in selection	Document inclusion criteria
No gap identification	Explicitly state what's missing

Quality Assessment Tools

Study Type	Assessment Tool
RCTs	Cochrane Risk of Bias (RoB 2)
Observational	Newcastle-Ottawa Scale
Systematic reviews	AMSTAR 2
Diagnostic studies	QUADAS-2
Qualitative	CASP Checklist

Integration with Grant System

Use aggregated grant data to:

1. Identify funded research: What topics have been funded recently?
2. Find gaps: What areas lack funding despite literature need?
3. Track trends: How are research priorities shifting?
4. Support significance: Cite funded grants as evidence of importance