Skip to main content
. 2023 Sep 8;2023(9):MR000054. doi: 10.1002/14651858.MR000054.pub2

Boynton 1998.

Study characteristics
Methods Objective
To develop a search filter using word frequency analysis ('objective approach') tested on the MEDLINE(Ovid). This method was validated with the use of expert input using the word frequency word list produced by ListIndex.
Data Quasi‐gold standard: produced by using the Institute for Scientific Information (ISI) citation index impact factors to identify six journals from the top ten high‐impact factor journals in the ‘Medicine general and internal’ category:

  • Annals of Internal Medicine

  • Archives of Internal Medicine

  • BMJ

  • JAMA

  • Lancet

  • New England Journal of Medicine


Hand search from 1992 to 1995 to identify systematic reviews and search via MEDLINE(Ovid).
Comparisons External validation: this filter was not externally validated
Comparisons to other filters (non‐validated): D.L. Hunt and K.A. McKibbon full and brief strategy ( Hunt 1997 and Centre for Reviews and Dissemination (CRD) brief and full strategy ( NHS 1996 )
Outcomes Sensitivity and precision (specificity could not be assessed due to the use of a quasi‐gold standard)
Notes Objectively derived filters
Strategy A:
review.pt.
Strategy B:
review.pt. 
data.ab. 
review tutorial.pt.* 
studies.ab. 
review.ab. 
clinical.ab. 
trials.ab. 
medline.ab. 
sources.ab. 
selection.ab.
Strategy C:
review.pt. 
review.ab.
trials.ab. 
medline.ab. 
sources.ab. 
selection.ab.
Strategy D:
medline.ab. 
sources.ab. 
selection.ab.
Strategy E:
medline.ab
Strategy F:
meta.ti. 
meta‐analysis.pt.* 
medline.ab. 
meta.ab. 
extraction.ab. 
english.ab. 
synthesis.ab. 
articles.ab. 
search.ab. 
language.ab. 
selection.ab. 
literature.ab. 
review academic.pt.* 
sources.ab.
Strategy H:
meta.ti. 
meta‐analysis.pt. 
medline.ab. 
meta.ab. 
extraction.ab. 
english.ab. review.pt. 
synthesis.ab. 
articles.ab. 
search.ab. 
language.ab. 
selection.ab. 
literature.ab. 
review academic.pt. 
sources.ab. 
randomized.hw. 
controlled.hw. 
published.ab. 
trials.ab. 
analysis.ti. 
relevant.ab. 
trials.hw. 
reviewed.ab. 
review.ab. 
review.pt.
Strategy J:
meta.ab. 
synthesis.ab. 
literature.ab. 
randomized.hw. 
published.ab. 
meta‐analysis.pt. 
extraction.ab. 
trials.hw. 
controlled.hw. 
search.ab.
medline.ab.
selection.ab.
sources.ab.
trials.ab.
review.ab.
review.pt.
articles.ab.
reviewed.ab.
english.ab.
language.ab.
Filters developed with the input of experts
Strategy K:
systematic adj 
review$.tw. 
data adj synthesis.tw. 
published adj studies.ab. 
data adj extraction.ab. 
meta‐analysis/ 
meta‐analysis.ti.
Strategy L:
systematic adj review$.tw.
data adj synthesis.tw. 
published adj studies.ab. 
data adj extraction.ab. 
meta‐analysis/ 
meta‐analysis.ti.
review.pt.
Strategy M:
meta‐analy$.ab.
data adj synthesis.ab.
literature adj3 review.ab.
randomized controlled trials/
published.ab.
meta‐analysis.pt.
data adj extraction.ab.
articles.ab.
reviewed.ab.
search.ab.
english adj language.ab.
All strategies excluded comments, letters, editorials and animal studies using the following strategy:
1 comment.pt. 
2 letter.pt. 
3 editorial.pt. 
4 animal/ 
5 human/ 
6 4 not (4 and 5) 
7 {SELECTED SEARCH STRATEGY} 
8 7 not (1 or 2 or 3 or 6)