• YOPL

PubMed Research Methodology Search Filters (and a couple of nifty limits)

Updated: Jun 15


Your search results are only as good as your search. Use a filter to separate the wheat from the chaff!


There will be a separate blog that covers subject oriented filters, such as vaccinations, age groups, etc. Stay tuned!


Be sure you take a look at the blog which gives an overview of search filters/search hedges. It can be found at: http://bit.ly/about-search-filters


As a reminder, we like to develop and run the subject content of the search strategy, then add the search filter and combine. Any additional limits, such as language or publication dates, would follow. Here is an example using the clinical trial search filter:


This also shows the magic of using and. Line 3 returned over 446,600 citations; line 6 returned over 2.72 million! Combining them with and reduced the number to a little over 43,200. Adding in the search filter, then limiting by language and date returns a little under 4,900 citations for potential clinical trials.


Following are research methodology search filters for the National Library of Medicine PubMed database. You cannot use these in other databases unless you first adapt the vocabulary to the database and the syntax to the interface!


Systematic Reviews & Meta-Analysis (updated 4/16/2019)

"Systematic Review"[PTYP] OR systematic critical review[tiab] OR systematic literature review[tiab] OR systematic literature search[tiab] OR systematic review[tiab] OR "systematic review of systematic reviews"[tiab] OR systematic reviews[tiab] OR systematic search[tiab] OR ((comprehensive[tiab] or systematic[tiab]) AND (bibliographic review[tiab] OR literature review[tiab] OR literature search[tiab] OR literature searches[tiab] OR narrative review[tiab])) OR comprehensive review[tiab] OR comprehensive search[tiab] OR evidence assessment[tiab] OR evidence review[tiab] OR integrative literature[tiab] OR integrative review[tiab] OR integrative reviews[tiab] OR (mapping[tiab] AND literature review[tiab]) OR mapping review[tiab] OR meta-review[tiab] OR rapid literature review[tiab] OR rapid literature search[tiab] OR rapid review[tiab] OR rapid reviews[tiab] OR scoping literature[tiab] OR scoping review[tiab] OR scoping reviews[tiab] OR scoping study[tiab] OR systematic narrative review[tiab] OR systematic qualitative review[tiab] OR systematic quantitative review[tiab] OR systematical literature review[tiab] OR systematical review[tiab] OR systematical reviews[tiab] OR umbrella review[tiab] OR umbrella reviews[tiab] OR cinahl[TIAB] OR cochrane[TIAB] OR embase[TIAB] OR medline[TIAB] OR psyclit[TIAB] OR (psycinfo[TIAB] NOT "psycinfo database"[TIAB]) OR pubmed[TIAB] OR scopus[TIAB] OR "sociological abstracts"[TIAB] OR "web of science"[TIAB] OR "Cochrane Database Syst Rev"[JO] OR "Evid Rep Technol Assess (Full Rep)"[JO] OR "Evid Rep Technol Assess (Summ)"[JO] OR "meta-analysis as topic"[MESH:noexp] OR Meta-Analysis[PTYP] OR "network meta-analysis"[mesh:noexp] OR meta analyses[ot] OR meta analyses[TIAB] OR meta analysis[TIAB] OR meta analytic[TIAB] OR meta analytical[TIAB] OR meta analytics[TIAB] OR meta analyze[TIAB] OR meta analyzed[TIAB] OR metaanalyses[TIAB] OR metaanalysis[ot] OR metaanalysis[TIAB] OR metaanalytic[TIAB] OR metaanalyze[TIAB] OR metaanalyzed[TIAB]

Note: Please contact me if you are interested in how I developed and tested this search filter.


The old search (4/1/2019; includes state of the art reviews, research synthesis, other terms; lots of false positives)

"Systematic Review" [PTYP] OR ((comprehensive*[TIAB] OR integrative[TIAB] OR mapping[TIAB] OR rapid[TIAB] OR scoping[TIAB] OR systematic[tiab] OR systematical[tiab] OR systematicallly[tiab] OR systematically[tiab] OR systematicaly[tiab] OR systematicly[tiab] OR systematicreview[tiab] OR systematicreviews[tiab] OR umbrella[TIAB]) AND (bibliographic[TIAB] OR bibliographical[TIAB] OR bibliographically[TIAB] OR bibliographics[TIAB] OR literature[TIAB] OR review[TIAB] OR reviews[TIAB])) OR "state of the art review"[TIAB] OR "research synthesis"[TIAB] OR ((((information[TIAB] OR data[TIAB]) and synthesis[TIAB]) OR (data[TIAB] AND (extract[TIAB] OR extracting[TIAB] OR extractings[TIAB] OR extractioin[TIAB] OR extraction[TIAB]))) AND (review[TIAB] OR review[PTYP])) OR (cinahl[TIAB] OR (cochrane[TIAB] AND (trial[TIAB] OR trials[TIAB]) OR embase[TIAB] OR medline[TIAB] OR psyclit[TIAB] OR (psycinfo[TIAB] not "psycinfo database"[TIAB]) OR pubmed[TIAB] OR scopus[TIAB] OR "sociological abstracts"[TIAB] OR "web of science"[TIAB]) OR "Cochrane Database Syst Rev"[JO] OR "Evid Rep Technol Assess (Full Rep)"[JO] OR "Evid Rep Technol Assess (Summ)"[JO] OR (review[TIAB] AND (rationale[TIAB] OR evidence[TIAB]) AND review[PTYP) OR meta analyses[TIAB] OR meta analysis[TIAB] OR meta analytic[TIAB] OR meta analytics[TIAB] OR meta analyze[TIAB] OR meta analyzed[TIAB] OR metaanalyses[TIAB] OR metaanalysis[TIAB] OR metaanalytic[TIAB] OR metaanalytics[TIAB] OR metaanalyze[TIAB] OR metaanalyzed[TIAB] OR "meta-analysis as topic"[MESH:NOEXP] OR Meta-Analysis[PTYP] OR "network meta-analysis"[mesh:noexp]


Just Cochrane reviews

"Cochrane Database Syst Rev"[JO]


Just Meta-Analysis (updated 4/15/2019)

"meta-analysis as topic"[MESH:NOEXP] OR Meta-Analysis[PTYP] OR "network meta-analysis"[mesh:noexp] OR meta analyses[ot] OR meta analyses[TIAB] OR meta analysis[ot] OR meta analysis[TIAB] OR meta analytic[ot] OR meta analytic[TIAB] OR meta analytical[ot] OR meta analytical[TIAB] OR (meta[ot] AND analytics[ot]) OR meta analytics[TIAB] OR meta analyze[ot] OR meta analyze[TIAB] OR (meta[ot] AND analyzed[ot]) OR meta analyzed[TIAB] OR metaanalyses[ot] OR metaanalyses[TIAB] OR metaanalysis[ot] OR metaanalysis[TIAB] OR metaanalytic[TIAB] OR metaanalyze[TIAB] OR metaanalyzed[TIAB]


Just Network Meta-Analysis (updated 4/16/2019)

"network meta-analysis"[mesh:noexp] OR "network meta analyses"[TIAB] OR network meta analysis[ot] OR network meta analysis[TIAB] OR indirect treatment comparison[ot] OR indirect treatment comparison[tiab] OR ((((mixed[ot] OR mixed[tiab] OR indirect[ot] OR indirect[tiab] OR indirectly[ot] OR indirectly[tiab]) AND (treatment[ot] OR treatment[tiab] OR treatments[ot] OR treatments[tiab] OR intervention[ot] OR intervention[tiab] OR interventions[ot] OR interventions[tiab] OR therapeutic[ot] OR therapeutic[tiab] OR therapeutics[ot] OR therapeutics[tiab]) AND (comparison[ot] OR comparison[TIAB] OR comparisons[ot] OR comparisons[TIAB])) OR mixed treatment[ot] OR mixed treatment[TIAB] OR mixed treatments[ot] OR mixed treatments[tiab] OR multiple treatment[ot] OR multiple treatment[TIAB] OR multiple treatments[ot] OR multiple treatments[TIAB]) AND ("meta-analysis as topic"[MESH:noexp] OR Meta-Analysis[PTYP] OR (Bayesian[ot] AND statistical[ot]) OR (Bayesian[tiab] AND statistical[tiab]) OR Bayesian statistics[ot] OR Bayesian statistics[tiab] OR meta analyses[ot] OR meta analyses[TIAB] OR meta analysis[ot] OR meta analysis[TIAB] OR meta analytic[ot] OR meta analytic[TIAB] OR meta analytical[ot] OR meta analytical[TIAB] OR (meta[ot] AND analytics[ot]) OR meta analytics[TIAB] OR meta analyze[ot] OR meta analyze[TIAB] OR (meta[ot] AND analyzed[ot]) OR meta analyzed[TIAB] OR metaanalyses[ot] OR metaanalyses[TIAB] OR metaanalysis[ot] OR metaanalysis[TIAB] OR metaanalytic[TIAB] OR metaanalytical[tiab] OR metaanalyze[TIAB] OR metaanalyzed[TIAB]))


Publication type=Review

Incredibly enough, not all systematic reviews have the publication type "review"! This will find articles that are general reviews of the literature, though.

review[PTYP]


Retractions, Withdrawals, Corrections, Comments, Replies to comments (10/18/2018)

"retraction of publication"[PTYP] OR Published Erratum[PTYP OR ((withdrawal[ti] OR retraction[ti]) AND (note[ti] or notice[ti])) OR comment[PTYP] OR reply[ti] OR comment[ti] OR comments[ti]

Note: When completing a systematic review, use this search after article review and study selection but before coding. You will need to be creative as there is no consistency in how the record appears.

EX: Consider doing a search with the author's name as well as a search with keywords from the title & abstract, then combine with OR to view the full list.

  1. wakefield aj[au] AND ("retraction of publication"[PTYP] OR Published Erratum[PTYP] OR ((withdrawal[ti] OR retraction[ti]) AND (note[ti] or notice[ti])) OR comment[PTYP] OR replies[ti] OR reply[ti] OR comment[ti] OR comments[ti] )

  2. ((measles[tiab] OR mmr[tiab]) AND autism[tiab]) AND ("retraction of publication"[PTYP] OR Published Erratum[PTYP] OR ((withdrawal[ti] OR retraction[ti]) AND (note[ti] or notice[ti])) OR comment[PTYP] OR reply[ti] OR comment[ti] OR comments[ti])

  3. #1 OR #2

Also, be sure you check out RetractionWatch (https://retractionwatch.com/). "Tracking retractions as a window into the scientific process"


Clinical trials (updated 4/15/2019)

Clinical Trial" [PTYP:NoExp] OR "Adaptive Clinical Trial" [PTYP:NoExp] OR "clinical trial, phase i"[PTYP] OR "clinical trial, phase ii"[PTYP] OR "clinical trial, phase iii"[PTYP] OR "clinical trial, phase iv"[PTYP] OR "controlled clinical trial"[PTYP] OR "multicenter study"[PTYP] OR "randomized controlled trial"[PTYP] OR "Clinical Trials as Topic"[mesh:noexp] OR "clinical trials, phase i as topic"[MeSH:noexp] OR "clinical trials, phase ii as topic"[MeSH:noexp] OR "clinical trials, phase iii as topic"[MeSH:noexp] OR "clinical trials, phase iv as topic"[MeSH:noexp] OR "controlled clinical trials as topic"[MeSH:noexp] OR "Non-Randomized Controlled Trials as Topic"[MeSH:noexp] OR "randomized controlled trials as topic"[MeSH:noexp] OR "Equivalence Trials as Topic"[MeSH:noexp] OR "Intention to Treat Analysis"[MeSH:noexp] OR "Pragmatic Clinical Trials as Topic"[MeSH:noexp] OR "early termination of clinical trials"[MeSH:noexp] OR "multicenter studies as topic"[MeSH:noexp] OR “Double-Blind Method”[Mesh] OR "phase I"[tiab] OR "phase II"[tiab] OR "phase III"[tiab] OR "phase IV"[tiab] OR "phase 1"[tiab] OR "phase 2"[tiab] OR "phase 3"[tiab] OR "phase 4"[tiab] OR ((randomised[TIAB] OR randomized[TIAB]) AND (trial[TIAB] OR trials[tiab])) OR ((single[TIAB] OR double[TIAB] OR doubled[TIAB] OR triple[TIAB] OR tripled[TIAB] OR treble[TIAB] OR treble[TIAB]) AND (blind*[TIAB] OR mask*[TIAB])) OR ("4 arm"[tiab] OR "four arm"[tiab])


Comparative studies (updated 8/8/2018)

This is a tough filter to do. The scope note from NLM for this study type is: “Comparison of outcomes, results, responses, etc for different techniques, therapeutic approaches or other inputs”; figuring out the right terms is complicated. Consider this a high-yield start to a search filter with probably a somewhat lower sensitivity (we're guessing ~80%) and higher specificity.

"comparative study"[PT:NOEXP] OR comparative[TIAB] OR "compared to"[TIAB] OR comparing[TIAB] OR comparison[TIAB]


Program evaluation (updated 12/19/2019)

evaluation study[PT:NOEXP] OR "evaluation studies as topic"[MESH:NOEXP] OR "program evaluation"[MESH:NOEXP] OR validation study[PT:NOEXP] OR "validation studies as topic"[MESH:NOEXP] OR (pre-[TIAB] AND post-[TIAB]) OR (pretest[TIAB] AND posttest[TIAB]) OR (program*[TIAB] AND (evaluate[TIAB] OR evaluated[TIAB] OR evaluates[TIAB] OR evaluating[TIAB] OR evaluation[TIAB] OR evaluations[TIAB] OR evaluator[TIAB] OR evaluators[TIAB])) OR effectiveness[TIAB] OR intervention[TIAB]


Epidemiologic Studies

NOTE: A word of caution for the next set of filters: Authors do not always use the appropriate terminology when describing their epidemiologic studies. This is true for the above as well, but we have noticed that the following study types are more frequently not identified as such in their titles or citations. Not only that, but a retrospective study can be a type of cohort study or it can be a type of case-control study! Use these filters with some caution. They will locate the kids of study they should, but only when the author knows what kind of study they have done and has described it as such!


Case-control studies (reviewed 8/8/2018)

Case-Control Studies[MESH:NOEXP] OR Control Groups[MESH:NOEXP] OR Matched-Pair Analysis[MESH:NOEXP] OR (case*[TIAB] AND control*[TIAB]) OR (case[TIAB] AND comparison*[TIAB]) OR control group*[TIAB]


Cohort studies (updated 8/6/2018)

cohort studies[mesh:noexp] OR longitudinal studies[mesh:noexp] OR follow-up studies[mesh:noexp] OR prospective studies[mesh:noexp] OR retrospective studies[mesh:noexp] OR cohort[TIAB] OR longitudinal[TIAB] OR prospective[TIAB]


Cross sectional studies (updated November 26, 2019)

Cross-Sectional Studies[mesh:noexp] OR Prevalence[mesh:noexp] OR (association[tiab] and (studies[tiab] OR study[tiab])) OR cross-sectional[tiab] OR prevalence[tiab] OR transversal[tiab] OR (association[ti] OR associations[ti])


Below is a search for biomarkers and sepsis association and cross sectional studies.




Incidence studies (updated 8/78/2018)

Incidence[MESH:NOEXP] OR incidence[TIAB] OR incidences[TIAB]


Prevalence studies (updated 8/7/2018)

Prevalence[MESH:NOEXP] OR cross-sectional studies[MESH:NOEXP] OR prevalence[TIAB] OR prevalences[TIAB]


Qualitative studies (updated 8/8/2018)

focus groups[MESH:NOEXP] OR "interviews as topic"[MESH:NOEXP] OR narration[MESH:NOEXP] OR qualitative research[MESH:NOEXP] OR (((face-to-face[TIAB] OR guided[TIAB] OR in-depth[TIAB] OR indepth[TIAB] OR semi-structured[TIAB] OR structured[TIAB] OR unstructured[TIAB]) AND (discussion*[TIAB] OR interview*[TIAB] OR questionnaire*[TIAB]))) OR ethnograph*[TIAB] OR "field work"[TIAB] OR fieldwork[TIAB] OR focus group*[TIAB] OR key informant*[TIAB] OR qualitative[TIAB]


Guidelines/Recommendations (updated 12/3/2015)

guidelines as topic[MeSH:noexp] OR practice guidelines as topic[MeSH:noexp] OR Health Planning Guidelines[MeSH:noexp] OR practice guideline[MeSH:noexp] OR clinical protocols[MeSH:noexp] OR Consensus[MeSH:noexp] OR "Consensus Development Conference"[PTYP] OR "Consensus Development Conference, NIH"[PTYP] OR "Consensus Development Conferences as Topic"[MeSH:noexp] OR "Consensus Development Conferences, NIH as Topic"[MeSH:noexp]or critical pathway[MeSH:noexp] OR (clinical[TIAB] AND pathway[TIAB]) OR (clinical[TIAB] AND pathways[TIAB]) OR (practice[TIAB] AND parameter[TIAB]) OR (practice[TIAB] AND parameters[TIAB]) OR algorithms[MeSH:noexp] OR care pathway[TIAB] OR care pathways[TIAB] OR guidance[TIAB] OR guideline*[TI]


Limit to US studies (updated Oct. 2, 2019)

Run your search as usual. Go to Advanced (under the search text boox). Copy the statement below and paste into the search text box on the advanced page. Fill in the last line number (there are 2 places to do this) of your search before hitting <Enter>.

((#last-line-number AND (north america[MESH:NOEXP] OR united states[MESH])) OR (#last-line-number NOT (africa[MESH] OR asia[MESH] OR australia[MESH] OR canada[MESH] OR central america[mesh] OR europe[MESH] OR south america[MESH] OR "Caribbean Region"[mesh:noexp] OR Aruba[mesh:noexp] OR "Caribbean Netherlands"[mesh:noexp] OR Curacao[mesh:noexp] OR "Sint Maarten"[mesh:noexp] OR "West Indies"[mesh:noexp] OR "Antigua and Barbuda"[mesh:noexp] OR Bahamas[mesh:noexp] OR Barbados[mesh:noexp] OR "British Virgin Islands"[mesh:noexp] OR Cuba[mesh:noexp] OR Dominica[mesh:noexp] OR "Dominican Republic"[mesh:noexp] OR Grenada[mesh:noexp] OR Guadeloupe[mesh:noexp] OR Haiti[mesh:noexp] OR Jamaica[mesh:noexp] OR Martinique[mesh:noexp] OR "Saint Kitts and Nevis"[mesh:noexp] OR Saint Lucia[mesh:noexp] OR "Saint Vincent and the Grenadines"[mesh:noexp] OR "Trinidad and Tobago"[mesh:noexp])))

NOTE: Be sure you put "#" in front of the last line number with no space.


Notice that in this case, the last search line before adding the US filter is line #7. The search filter was modified in 2 places-- in the first major parentheses statement and then in the 2nd major parentheses statement just before NOT (africa[mesh]....

Limit to Human Studies (November 26, 2019)

Avoid checking selecting the limit to human studies-- you are limiting to a MeSH term. The newest, most current citations do not have MeSH terms. Consequently, you are kicking out the most recent research by using the human limiter. The following method does not do that.


last-line-number NOT ("Animals"[mesh] NOT ("Animals"[mesh] AND "Humans"[mesh]))

NOTE: Be sure you put "#" in front of the last line number with no space.


Limit by age groups (updated April 20, 2020)

The following method removes a significant number of studies (but NOT all) by age groups-- but it won't be 100%.

Remove child AND adolescent studies, keep adult studies

last-line-number NOT ((child[mesh] OR adolescent[mesh]) NOT ((child[mesh] OR adolescent[mesh]) AND adult[mesh]))

Remove child studies, keep adolescent AND adult studies

last-line-number NOT (child[mesh] NOT (child[mesh] AND adult[mesh]))

Remove child, adolescent, AND adult studies, keep infant studies

last-line-number NOT ((child[mesh] OR adolescent[mesh] OR adult[mesh]) NOT ((child[mesh] OR adolescent[mesh] OR adult[mesh]) AND infant[mesh]))



  • Google+ Social Icon
  • Facebook Social Icon
  • LinkedIn Social Icon
  • Twitter Social Icon

© 2018 by Your Own Private Librarian LLC. Proudly created with Wix.com