Integrative Review Capstone Toolkit

The purpose of the database searching phase is to design and execute a literature search that allows you to retrieve records for literature that is potentially eligible to be included in your review

The purpose of the article selection phase is to identify the articles that you will appraise, analyze and synthesize in your review

Guidance

For published guidance about this phase, also see:

• Toronto & Remington, Chapter 3
  • Lawless, J., & Foster, M. J. (2020). Searching systematically and comprehensively. In C. E. Toronto & R. Remington (Eds.), A Step-by-Step Guide to Conducting an Integrative Review (pp. 21–44). Springer International Publishing. https://doi.org/10.1007/978-3-030-37504-1_3 
• Whittemore & Knafl, "Literature search stage" (pp. 548-549)
  • Whittemore, R., & Knafl, K. (2005). The integrative review: updated methodology. Journal of Advanced Nursing, 52(5), 546-553.

Foundational Skills & Knowledge Required

To design an effective literature search for an integrative review and identify studies for your review, you should understand:

• what a literature database is
• which databases cover the literature related to your topic of interest
• how to use Boolean operators, punctuation, keywords and subject headings to structure a search
• how to balance sensitivity and specificity in a search
• your criteria for including or excluding literature from your review

Planning Questions to Consider

• Based on your review topic, which databases should you use to locate literature?
• What are your core search criteria?
  • Which of your eligibility criteria will you build into the search queries ("search criteria"), and which criteria will you apply manually as you screen the resulting records ("screening criteria")?
• How will you structure your searches?
  • What terms will you use? 
  • What database filters make sense
• How will you document your search process? (save queries in databases? External documents?)
• How systematic are you going to be in the searching and screening process? For instance:
  • Less systematic: Screen records for title/abstract relevance in database, only save articles for full text review
  • More systematic: Export records from finalized search in database into citation manager, de-duplicate, screen titles/abstracts for relevance, screen full texts for inclusion, track exclusion reasons, report via PRISMA diagram

Core Activities

As you move through the literature searching phase, you will be:

• Selecting databases to search for literature
• Experimenting with broader and narrower search queries to find a balance of specificity and sensitivity
• Designing, refining and testing comprehensive search queries for each database you’re using 
• Finalizing search queries and ensuring queries are conceptually consistent across databases
• Submitting search queries for feedback from a librarian

As you move through the literature selection phase, you will be:

• Applying the review's eligibility criteria to screening the titles and abstracts of the records you locate in the databases to determine if they could represent eligible literature
• Applying the review's eligibility criteria to screen the full texts of potentially relevant literature to determine if the literature is eligible to include in your review

Goals & Outcomes

When the database searching phase is complete, you should have:

• A database search query for each database that captures a set of records for literature that could be potentially relevant 
• A relatively broad set of database records to select from during screening

When the article selection phase is complete, you should have:

• A set of full texts of articles that meet your eligibility criteria (address your review purpose/question) which you can then subject to critical appraisal, data extraction, data analysis and synthesis

From your database searches, you'll report:

• Which databases (on which platform) you searched to discover literature that meets your criteria
• How you structured your searches in each database (search criteria)
• The exact queries you used in each database (including Boolean logic, punctuation, and filters)
• How many records you located in each database with those queries

From your screening and article selection, you'll report:

• How many records did you exclude because they were irrelevant, based on the information in the title/abstract?
• How many articles did you need to read full text for in order to make an inclusion decision? 
• How many articles did you locate that are eligible to be included in your final synthesis?

Reporting with PRISMA

If you plan to report the results of the literature search and selection phase using a PRISMA diagram, you will also need to use a citation manager (e.g., Zotero or Covidence) to remove duplicate records from the database search results. You would export the results sets from each database and pool them in a citation manager before screening in order to report the number of unique records screened.  

For an example of what data to collect in order to be able to create a basic PRISMA diagram, see this document: Data Required for PRISMA Diagram

(You'll need to be logged in with your NYU Google Account for access, but you may make a copy for your own use) 

As you're embarking on a search for "literature", it might be helpful to take a step back and consider what might consist of "literature" for the health sciences.  

In the context of literature reviews, it is useful to think about "the literature" as a place of scholarly communication. 

In other words, "the literature" is the product of scholars, researchers, and practitioners creating knowledge.  In the literature, they share ideas, theories, discoveries and viewpoints on topics that matter for their discipline.  Typically, these products take the form of written articles or conference presentations (perhaps accompanied by posters or abstracts) in which scholars:

• report the results of empirical research studies
• propose a new theory or theoretical framework
• share a new methodology, or evaluate existing methodologies used in the field 
• present an expert opinion or commentary on a domain-relevant topic
• synthesize or summarize existing knowledge

The diagram below summarizes the types of information that scholars may share in the form of literature.

(Some) Types of Scholarly Literature in the Health Sciences

Research Findings

• Empirical Research
  • Quantitative Studies
    • Experimental Methods
      • Randomized controlled trials
      • Non-randomized controlled trials 
    • Observational Methods
      • Cohort study
      • Case-control study
      • Case report/series/study
    • Qualitative Studies
      • Phenomenological study
      • Ground theroy
      • Ethnographic study
• Theoretical Research
• Methodological Research

Synthesis of Research

• Literature Reviews
  • Narrative reviews
  • Systematic or systematic-style reviews
• Clinical / practice guidelines

Expert Commentary 

• Commentaries / opinion pieces
• Editorials
• Reviews of books/resources
• Letters to the Editor