Searching for Relevant Studies (Text Version)

Slide Presentation from the AHRQ 2009 Annual Conferenc

Slide presentation from the AHRQ 2009 conference.

On September 16, 2009, Michael White made this presentation at the 2009 Annual Conference. Select to access the PowerPoint® presentation (1 MB) (Plugin Software Help).

Slide 1


 

Searching for Relevant Studies

C. Michael White, Pharm.D., FCP, FCCP
Professor and Director.University of Connecticut / Hartford Hospital.Evidence-based Practice Center

Author has no actual or potential conflicts of interest in relation to this activity

 

Slide 2


 

Outline

  • This section will:
    • Substantiate why prudent literature searching is important
    • Describe important tenets of literature searching
    • Describe why transparency in the search process is important

 

Slide 3


 

Balancing Precision and Recall

Precision: Proportion of retrieved articles that are relevant
Recall: Proportion of potentially relevant articles retrieved by search

  • Systematic reviews require comprehensive searches
  • Searching many databases using a very generic term gives high recall but is very imprecise
    • Searching for cancer.mp or Neoplasms yields 786,978 citations
      • That is a lot of citations to search through to yield the 323 citations that mention the role of tamoxifen and raloxifene in cancer, the reason for your search
  • The goal is to carefully construct a search in the attempt to keep precision high while maximizing recall

AHRQ Methods Guide, Finding Evidence, Chapter 5.; White CM. Ovid SP Search July 23, 2009.

 

Slide 4


 

Presenting Popular Studies Not Enough

  • Studies published from 1993 to 2000 in high-impact factor journal with positive findings & cited >1000 times included (n=45)
  • Results compared against other trials with same comparators employing populations and methodologies that were similar or superior
  • Opposite/null findings or much more dramatic effects found 31% of the time
    • 83% of NR studies and 23% of RCTs

Ioannidis JPA. JAMA 2005;294:218-28.

 

Slide 5


 

Where to Begin

  • When a general topic is proposed it is tempting to begin by extensively searching for primary literature
    • "The effectiveness of generic versus innovator antiepileptic drugs in patients with epilepsy."
  • Understanding the topic, devising an analytic framework, asking clearly defined key questions, and understanding the scope of the review you wish to conduct is crucial before performing the extensive literature search

AHRQ Methods Guide, Finding Evidence, Chapter 5.; White CM. AHRQ Topic Refinement Report, Submitted 7.15.09.

 

Slide 6


 

Specialized Electronic Databases

AHRQ Methods Guide, Finding Evidence, Chapter 5.

Generally a minimum requirement for a thorough search of literature
 

  • While the topic area dictates the databases to be included, here are several common ones:
    • MEDLINE® (General)
    • Cochrane CENTRAL (General)
    • EMBASE (General, more international in scope)
    • PsychLIt/PsychINFO (Psychology)
    • AIDSLine (HIV/AIDS)
    • CINAHL (Nursing)
    • TOXNET (Adverse Events/Toxicology)

 

Slide 7


 

Medline Alone is Not Enough

  • Study 1: Sensitivity and specificity of using Medline vs. 9 databases
    • Ability to retrieve economic analyses from January to March 1997
    • Medline only searches had a sensitivity of 72% and specificity of 75% vs. 9 databases
  • Study 2: Systematic review of prevalence of maternal mortality and morbidity from 1997 to 2002
    • Multiple databases searched (Medline, EMBASE, BIOSIS, LILACS)
    • 60% of citations were found in >1 database
    • Medline search had 20% of non-replicated citations, EMBASE had 7.4%, LILACS 5.6%

Sassi F. Medical Care 2002;40:387-94.; Betran AP. BMC Med Res Methodol 2005, Vol 5, No 1, 6.

 

Slide 8


 

Pilot Searching

  • Perform a pilot search and look to see if articles already identified by the research group, Key Informants, and manual searches of references of these articles turn up in the search
    • Pilot searching usually limited to Medline and then adapted to Cochrane Central and possibly other databases
    • If the pilot search is inadequate, refine the search
    • Searching the literature is an iterative process

AHRQ Methods Guide, Finding Evidence, Chapter 5.

 

Slide 9


 

Multiple Searches

  • Multiple searches within the same database may be necessary
  • Efficacy and harms searches may be separate
    • Small size of available literature, single broad search is appropriate
    • Extensive size of available literature:
      • Efficacy search may include the disease, intervention, and randomized controlled trial search
        • Efficacy evaluations frequently are limited to RCTs
          • Fewer inherent biases, higher quality evidence
        • Harms search may be broader and include just the intervention and harms ± disease
          • Paucity of harms data in published RCTs frequently require non-RCTs be included
          • Authors would likely be interested in harms regardless of the reason for the intervention

AHRQ Methods Guide, Finding Evidence, Chapter 5.

 

Slide 10


 

Boolean Operators

  • "OR" makes the search broader
  • "AND" makes the search more selective
  • "adj" terms cut down on misc citations
    • Coronary adj artery (only shows up if coronary next to artery)
  • ".ti" or ".ab" searches in the title or abstract for the word of interest
    • Can find citations where the word you search for is not a keyword
  • "/" means all subheadings are searched
    • When a search term is entered, you can selectively choose some subheadings (such as pharmacology for pharmacology studies)
    • Subheadings generally imprecise, accepting all subheadings increases the yield
  • "$" truncates a word with different endings
    • Analy$ would pick up words like analysis, analyses, analyze, analyse
  • ".mp" is a text word search, doesn't need to be a keyword to show up

 

Slide 11


 

Harms Searching

AHRQ Methods Guide, Finding Evidence, Chapter 5.

  • If the specific adverse effects of interest are specified in the key questions, they can be searched for directly (rhabdomyolysis, cancer)
  • General harms subheadings for Medline would include:
    • /adverse effects
    • /poisoning
    • /toxicity
    • /chemically induced
    • /contraindications
    • /complications
  • General harms subheadings for EMBASE would include:
    • /side effect
    • /adverse drug reaction
    • /drug toxicity
    • /complication
  • Consider a database dedicated to harms such as TOXNET http://toxnet.nlm.nih.gov/index/html

 

Slide 12


 

Using Filters (or Hedges)

AHRQ Methods Guide, Finding Evidence, Chapter 5.

  • Filters have been developed and validated that can help balance precision and recall
    • Cochrane—www.cochrane-handbook.org/ section 6.4.11 Search Filters, National Health Service Centre for Reviews and Dissemination www.york.ac.uk.inst/crd/revs.htm, Scottish Intercollegiate Guidelines Network www.sign.ac.uk/methodology/filters/html, InterTASC Information Specialists' Sub-Group www.york.ac.uk/instintertasc/index/htm.
  • Filters may not work in all circumstances

 

Slide 13


 

Finding RCTs: Filter (Hedge)

Terms for Controlled Trials. .pt = Study Type, .ti = Title, .ab = Abstract, .sh = subheading

  1. randomized controlled trial.pt.
  2. Controlled clinical trial.pt.
  3. randomized.ab.
  4. Placebo.ab.
  5. Clinical trials as topic.sh.
  6. randomly.ab.
  7. Trial.ti.
  8. 1 or 2 or 3 or 4 or 5 or 6 or 7

Baker WL. AHRQ CER Report on ACE Inhibitors and ARBs.


 

Slide 14


 

Finding Observational Studies: Filter (Hedge)

All Terms for Observational Studies. .tw = Term in Title and Abstract, adj = Adjacent, / = Accepted All Subheadings, $ = Truncated Word, exp = Exploded Term

  • MEDLINE (OVID) for Observational Studies using the Scottish Intercollegiate Guidelines Network Observational Study MEDLINE Search Filter (available at: http://www.sign.ac.uk/methodology/filters.html)
    1. Epidemiologic studies/
    2. Exp case control studies/
    3. Exp Cohort Studies/
    4. Case control.tw.
    5. (cohort adj (study or studies)).tw.
    6. Cohort analy$.tw.
    7. (follow up adj (study or studies)).tw.
    8. (observational adj (study or studies)).tw.
    9. Longitudinal.tw.
    10. retrospective.tw.
    11. Cross sectional.tw.
    12. Cross-Sectional Studies/
    13. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12

Baker WL. AHRQ CER Report on ACE Inhibitors and ARBs.

 

Slide 15


 

PRESS Checklist for Searches

  • Items to consider before hitting enter:
    • Spelling errors/spelling in American and English
      • Anaesthesia and anesthesia, analyse and analyze, oestrogen and estrogen
    • Line errors (when searches are combined using line numbers. i.e. "1 AND 3" instead of "1 AND 4"
    • Boolean operators used appropriately
    • Search strategies adapted for specific databases
    • All appropriate headings used (refer to thesaurus for specific database)
    • Appropriate use of subheadings and floating subheadings

AHRQ Methods Guide, Finding Evidence, Chapter 5.

 

Slide 16


 

PRESS Checklist for Searches—continued

  • Items to consider before hitting enter:
    • Use of natural language terms (search as key word)
      • Index terms are great but there is a lag time for indexing, inappropriate indexing, lack of adequate indexing terms, changes in terms over time without retrospective updating
    • Truncation and spelling variation as appropriate
    • Appropriate use of limits such as language, years
    • Field searching, publication type, author

AHRQ Methods Guide, Finding Evidence, Chapter 5.

 

Slide 17


 

Publication Bias

AHRQ Methods Guide, Finding Evidence, Chapter 5.

  • Publication bias is the tendency of certain types of trials (such as those with the largest effects) to be published
  • Publication bias increases the risk that the observed effect might not reflect the true effect
    • May negatively impact consistency, precision, and magnitude of effect
  • Expanding searches to include additional languages, citation tracking, hand searching, and grey literature can help identify and possibly minimize publication bias

 

Slide 18


 

Publication Bias Example

Turner EH. NEJM 2008;358:252-60.

  • Study: 74 antidepressant studies registered with the FDA
    • 97% of positive studies published
    • 39% of neutral or negative studies published
      • 11 of 14 published in a way that conveyed the positive results but deemphasized the negative
    • When only published literature was meta-analyzed, a 32% increase in relative effect size occurred versus the more complete dataset of conducted trials

 

Slide 19


 

Language Restrictions

  • It is not a requirement to use foreign language publications
    • EPCs should consider how inclusion or exclusion of foreign language studies might or might not bias the search
    • Usually advisable to include foreign language studies for complementary and alternative medicine topics

AHRQ Methods Guide, Finding Evidence, Chapter 5.

 

Slide 20


 

English Only or All Languages

  • Study 1: Evaluation of 42 systematic reviews including 662 RCTs
    • For traditional medicine, language-restricted systematic reviews, compared with language-inclusive ones, did not introduce biased results, in terms of estimates of intervention effectiveness (ROR=1.02; 95% CI=0.83-1.26)
    • For CAM, however, language-restricted systematic reviews resulted in a 63% smaller protective effect estimate than language-inclusive reviews (ROR=1.63; 95% CI=1.03-2.60).

Pham B. J Clin Epidemiol 2005;58:769-76.

 

Slide 21


 

English Only or All Languages

  • Study 2: Evaluation of acupuncture trials conducted in England vs. China, Japan, Russia, Taiwan
    • RCTs or controlled trials in England were favorable of acupuncture 75% of the time
    • Results were favorable 99%, 89%, 97%, and 95% of the time in China, Japan, Russia, and Taiwan, respectively
    • Researchers should consider carefully how to manage data from these countries

Vickers A. Controlled Clin Trials 1998;19:159-66.

 

Slide 22


 

Citation Tracking

  • Forward citation analysis: these databases provide information on articles cited by other authors
    • Relies on author's own choice to cite an article rather than keywords or indexing
      • As such, can turn up unique items and can complement a traditional database search
    • Web of Science, Scopus, Google Scholar, PubReMiner, PubFocus
      • Web of Science and Scopus have access fees but have most developed search and export interfaces
      • Google Scholar, like the other remaining ones, are free access but not as developed

AHRQ Methods Guide, Finding Evidence, Chapter 5.

 

Slide 23


 

Citation Tracking—Continued

AHRQ Methods Guide, Finding Evidence, Chapter 5.

  • Backward citations: Reading references of key articles
    • New search terms, poorly indexed journals, or inappropriate limiters in the original search can be identified
  • "Related article" links: Useful when designing and refining a search but not for formal search
    • Difficult to perform systematically, report strategy, and reproduce the search

 

Slide 24


 

Hand Searching

AHRQ Methods Guide, Finding Evidence, Chapter 5.

  • Not all journals may be indexed by databases you searched
    • Some journals only index certain article types
    • Abstracts, conference proceedings, and supplements may not be indexed
    • Backward citation searching and Key Informant discussion can identify these journals
    • Ulrich has a useful website to identify journals in a specific topic area (www.ulrichweb.com/ulrichsweb/)
      • If Ulrich indicates an important journal is not indexed fully in databases you use, search by hand

 

Slide 25


 

Hand Searching Benefits

Hopewell S. Stat Med 2002;21:1625-34. Jadad AR. Online J Curr Clin Trials Vol. Doc No 33 (1993)

  • Study 1: Using Medline to search 22 specialized journals to identify RCTs was compared with hand searching the journals
    • 117 of the 462 randomized controlled trials (25%) were not retrieved by Medline due to a lack of publication type terms 'randomized controlled trial' or 'controlled clinical trial'
      • Additional articles were missed in journals without full journal indexing
  • Study 2: Using Medline to search 9 pain and anesthetic journals as compared with handsearching
    • 13% of eligible articles were not retrieved with Medline alone

 

Slide 26


 

Grey Literature

  • Grey literature: Literature produced on all levels of government, academics, business, and industry not controlled by commercial publishers
    • Usually not systematically identified, stored, or indexed
    • Can help overcome publication bias: tendency of authors, journals, or reviewers to preferentially publish positive studies
  • Least efficient type of literature to search and may be (or are perceived to be) of lower quality
    • Useful in areas with little published evidence, field that is rapidly changing (devices, surgery), or field that is highly interdisciplinary (mental health or alternative medicine)

AHRQ Methods Guide, Finding Evidence, Chapter 5.

 

Slide 27


 

Grey Literature

Hopewell S. Cochrane Database of Systematic reviews, No. 2, 2007. Turner EH. NEJM 2008;358:252-60.

  • Study: Reviewed 5 systematic reviews conducted to evaluate the impact of grey literature in meta-analyses of randomized controlled trials
    • All systematic reviews found more positive results with published literature than grey literature (ROR 1.09; 95% CI 1.03-1.16)

 

Slide 28


 

Grey Literature: Regulatory Search

  • Approval process for new drugs and devices involves submissions of data which may not be published or not completely published elsewhere
    • Good check for publication bias
  • Drugs@FDA
    www.accessdata.fda.gov/Scripts/cder/DrugsatFDA/index.cfm?fuseaction=Search.search_Drug_Name
  • FDA Center for Drug Evaluation and Research (CDER) Archives
    www.fda.gov/cder/archives/default.htm
  • Dockets for FDA
    www.fda.gov/ohrms/dockets/default.htm
  • Devices@FDA
    www.accessdata.fda.gov/Scripts/cdrh/devicesatFDA/
  • Health Canada Drug Product Search
    http:205.193.93.51/dpdonline/startup.do?applanguage=en_CA
  • European Medicines Agencies EPARs for Authorized Medicinal Products for Human Use
    www.emea.europa.eu/htms/human/epar/a.htm

AHRQ Methods Guide, Finding Evidence, Chapter 5.

 

Slide 29


 

Grey Literature: Trial Registries

AHRQ Methods Guide, Finding Evidence, Chapter 5.

  • Online registries include trial name, type, and brief methodology
    • May contain results of completed but unpublished trials as well
  • ClinicalTrials.gov clinicaltrials.gov/ct2/search
  • ClinicalStudyResults.org www.clinicalstudyresults.org/home/
  • Current Controlled Trials www.controlled-trials.com/mrct/
  • Australian New Zealand Clinical Trials Registry www.anzctr.org.au/trialSearch.aspx
  • Netherlands Trial Registry www.trialregister.nl

 

Slide 30


 

Grey Literature: Abstracts & Conference Proceedings

AHRQ Methods Guide, Finding Evidence, Chapter 5.

  • Much of this literature unlikely to be published
  • When published, final data often differs from that in the abstract
  • Use Key Informants to identify the most important meetings in the field of interest and hand search the abstracts and conference proceeding supplements of their journals
    • Some databases are available as well: ProceedingsFirst, Biological Abstracts, Conference Papers Index, Conference Proceedings Citation Index (part of Thompson Reuters Web of Science), BioWizard Scholar Conference Abstract Search

 

Slide 31


 

Directly Contacting Researchers

  • Researchers can be contacted to determine if projects in grey literature are published in a peer reviewed form
  • Prolific researchers in the field can be contacted to see if they know of any citations that were missed by the literature search
  • Search engine (Yahoo, Google, etc) queries with the individuals name or institution can be a good place to start
    • Contact information on manuscripts also a reasonable place to start

AHRQ Methods Guide, Finding Evidence, Chapter 5.

 

Slide 32


 

Reporting Search Strategies: Transparency

AHRQ Methods Guide, Finding Evidence, Chapter 5.

  • Accurate and transparent reporting of search strategies is important
    • Searches need to be updated towards the conclusion of your CER to assure that no new information has come out in the intervening period
    • Stakeholders and end users want assurances that the CER was free of bias and that adequate search principles were followed
    • Searches will need to be re-run if the CER is updated several years after the CER is completed

 

Slide 33


 

Transparent Reporting

AHRQ Methods Guide, Finding Evidence, Chapter 5.

  • Databases used
  • Dates covered
  • Search terms
  • Language restrictions
  • Non-database methods used
  • Inclusion/exclusion criteria
  • Full electronic search strategy
  • Publication related restrictions
  • End date of search
  • List of excluded references
  • Qualifications of searcher
  • Number of references identified
  • CONSORT or QUOROM - style flow diagram accounting for all references
  • Evidence of search effectiveness
  • Statement of filters employed
  • Description of sampling strategy

 

Slide 34


 

Transparent Reporting: QUOROM Figure

Figure: An example of a controlled trial QUOROM Figure.

 

Slide 35


 

Conclusions

  • Searching is important: balance precision and recall
    • Medline is not enough
  • Multiple strategies improve completeness of search
  • Hedges are a good place to start
  • You may need more than one search (efficacy and harms)
  • Report strategy with transparency

 

Slide 36


 

Free Bonus Material!!!

  • The following slides has additional bonus material for you to review
    • Additional grey literature sources
    • Scientific information packets
    • Full example of search within an EPC report
      • See how the search pieces are created and then comes together
        • Note the use of Boolean operators

 

Slide 37


 

Grey Literature: Theses & Dissertations

  • Becoming increasingly available as institutions post them online
  • Databases: ProQuest Dissertation & Theses (the most comprehensive collection in the US with full text of dissertations through UMI's Digital Archiving and Access program), Index to Theses in Great Britain and Ireland, Networked Digital Library of Theses and Dissertations (NDLTD), Dissertation Abstracts Online

AHRQ Methods Guide, Finding Evidence, Chapter 5.

 

Slide 38


 

Grey Literature: Government Documents

  • Aside from governmental regulatory sites, other governmental agencies support or collect grey literature
  • Computer Retrieval of Information on Scientific Projects (CRISP) crisp.cit.nih.gov - searchable database of federally funded biomedical research
  • Health Services Research Projects in Progress (HSRPROJ) www.cf.nlm.nih.gov/hsr_project/home_proj.cfm—database of ongoing grants and projects in health services research
  • Science Accelerator www.scienceaccelerator.gov/ from the Office of Scientific and Technical Information and the US Department of Energy
  • Defense Technical Information Center (DTIC) www.dtic.mil/dtic/ from the Department of Defense including the military health system

 

Slide 39


 

Grey Literature: General Sources

  • Grey Matters: a practical tool for evidence-based searching cadth.ca/index.php/en/cadth/products
    • Collated list of sources for grey literature appropriate for health sciences by the Canadian Agency for Drugs and Technology in health (CADTH)
  • System for Information on Grey Literature in Europe (SIGLE) opensigle.inist.fr/
    • Provides bibliographic references to grey literature in Europe
  • New York Academy of Medicine (NYAM) www.nyam.org/library.grey.shtml
    • Provides their grey literature on the website with a list of other organizations that do the same

AHRQ Methods Guide, Finding Evidence, Chapter 5.

 

Slide 40


 

Scientific Information Packets

AHRQ Methods Guide, Finding Evidence, Chapter 5.

  • Provided by the company manufacturing the drug or device
  • SIPs include information about products including the product label, published, and unpublished trials or studies
    • Can help to overcome publication bias by identifying trials that remain unpublished
    • Soon to be released trials can be identified and ultimately captured in CERs
    • Allows drug and device manufacturers to be explicitly involved in helping provide a literature base to a CER

 

Slide 41


 

Scientific Information Packets

A flow chart appears on the slide depicting the following order of steps:

  • EPC Determines The Drug or Devices Covered in CER
  • SRC Takes List and Identifies Companies (Uses FDA Website if Companies Unknown)
  • SRC Identifies Contacts (Uses FDA Website, Company Website, Hoovers http://www.hoovers.com/free/ or Gale's Business & Company Resource Center http://www.gale.cengage.com/businessRC/index.htm)
  • SRC Contacts Via Letter and e-mail. Explains Nature of Review, What is Requested (Published and Unpublished Trials or Protocols, Registry Information), Identifies That Information Could Be Publicly Disclosed Via FOIA
  • SRC Transmits Information to EPC

Source: AHRQ Methods Guide, Finding Evidence, Chapter 5.
 

 

Slide 42


 

Example of Controlled Trial Search

MEDLINE (OVID) for Randomized Controlled Trials Using the Cochrane Highly Sensitive and Specific Search Strategy (Sensitivity and Precision Maximizing Version 2008)

Terms for Vascular Diseases. / = All Subheadings Were Selected

  • 1. Coronary Artery Disease/ or Coronary Disease/
  • 2. Myocardial Ischemia/
  • 3. Angina Pectoris/ or Angina, Unstable/
  • 4. Angina Pectoris/ or Arterial Occlusive Diseases/
  • 5. Peripheral Vascular Diseases/
  • 6. Vascular Diseases/
  • 7. Atherosclerosis/
  • 8. Cardiovascular Diseases/
  • 9. Carotid Artery Diseases/

Terms for Preserved.or Stable Disease..Adj = Adjacent

  • 10. (((preserved adj left) or (stable adj cad) or (stable adj chd) or or (stable adj coronary) or (preserved adj coronary) or (preserved adj systolic) or (preserved adj ventricular) or (preserved adj lvef) or (preserved adj ef) or (preserved adj ejection)) or (intact adj left) or (intact adj systolic) or (intact adj ventricular) or (intact adj lvef) or (intact adj ef) or (normal adj systolic) or (normal adj ventricular) or (normal adj lvef) or (normal adj ef)).mp.
  • 11. 1 or 23 or 4 or 5 or 6 or 7 or 8 or 9 or 10

 

Slide 43


 

Example of Controlled Trial Search

Terms for Controlled Trials. .pt = Study Type, .ti = Title, .ab = Abstract

  • 12. randomized controlled trial.pt.
  • 13. controlled clinical trial.pt.
  • 14. randomized.ab.
  • 15. placebo.ab.
  • 16. clinical trials as topic.sh.
  • 17. randomly.ab.
  • 18. trial.ti.
  • 19. 12 or 13 or 14 or 15 or 16 or 17 or 18

Baker WL. AHRQ CER Report on ACE Inhibitors and ARBs.

 

Slide 44


 

Example of Controlled Trial Search

  • 20. humans.sh.
  • 21. 19 and 20

All Headings for ACE inhibitors and ARBs Including Free Text (.mp)

  • 22. (alacepril or benazepril or captopril or ceronapril or cilazapril or delapril or enalapril or fosinopril or imidapril or libenzapril or lisinopril or moexipril or moveltipril or pentopril or perindopril or quinapril or ramipril or spirapril or temocapril or teprotide or trandolapril or zofenopril).mp.
  • 23. (losartan or olmesartan or telmisartan or valsartan or eprosartan or candesartan or tasosartan or irbesartan).mp.
  • 24. Angiotensin-Converting Enzyme Inhibitors/
  • 25. Angiotensin II Type 1 Receptor Blockers/
  • 26. (ACEI or ARB).mp.
  • 27. 22 or 23 or 24 or 25 or 26

All Vascular Disease Studies In Humans Who Are.Stable And Have Preserved Ventricular Function.Evaluating ACE inhibitors or ARBs

  • 28. 11 and 21 and 27

Baker WL. AHRQ CER Report on ACE Inhibitors and ARBs.

 

Slide 45


 

Example of Observational Study Search

All Terms for Observational Studies. .tw = Term in Title and Abstract, adj = Adjacent, / = Accepted All Subheadings, $ = Truncated Word, exp = Exploded Term

  • MEDLINE (OVID) for Observational Studies using the Scottish Intercollegiate Guidelines Network Observational Study MEDLINE Search Filter (available at: http://www.sign.ac.uk/methodology/filters.html)
  • 1. epidemiologic studies/
  • 2. exp case control studies/
  • 3. exp Cohort Studies/
  • 4. case control.tw.
  • 5. (cohort adj (study or studies)).tw.
  • 6. cohort analy$.tw.
  • 7. (follow up adj (study or studies)).tw.
  • 8. (observational adj (study or studies)).tw.
  • 9. longitudinal.tw.
  • 10. retrospective.tw.
  • 11. cross sectional.tw.
  • 12. Cross-Sectional Studies/
  • 13. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12

Baker WL. AHRQ CER Report on ACE Inhibitors and ARBs.

 

Slide 46


 

Example of Observational Study Search

All Terms for ACE Inhibitors and ARBs. .mp = Free Text Term, / = All Subheadings Accepted

  • 14. (alacepril or benazepril or captopril orceronapril or cilazapril or delapril or enalapril or fosinopril or imidapril or libenzapril orlisinopril or moexipril).mp.
  • 15. (moveltipril or pentopril or perindopril or quinapril or ramipril or spirapril or temocapril or teprotide or trandolapril or zofenopril).mp.
  • 16. (losartan or olmesartan or telmisartan or valsartan or eprosartan or candesartan or tasosartan or irbesartan).mp.
  • 17. Angiotensin-Converting Enzyme Inhibitors/
  • 18. Angiotensin II Type 1 Receptor Blockers/
  • 19. (ACEI or ARB).mp.
  • 20. 14 or 15 or 16 or 17 or 18 or 19

Baker WL. AHRQ CER Report on ACE Inhibitors and ARBs.

 

Slide 47


 

Example of Observational Study Search

All Search Terms for Preserved and Stable Disease. .Adj = Adjacent, ..mp = Free Text Word

  • 21. (((preserved adj left) or (stable adj cad) or (stable adj chd) or (stable adj coronary) or (preserved adj coronary) or (preserved adj systolic) or (preserved adj ventricular) or (preserved adj lvef) or (preserved adj ef) or (preserved adj ejection)) or (intact adj left) or (intact adj systolic) or (intact adj ventricular) or (intact adj lvef) or (intact adj ef) or (normal adj systolic) or (normal adj ventricular) or (normal adj lvef) or (normal adj ef)).mp

Search for Studies of Observational Trials Evaluating ACE inhibitors or ARBs in Preserved or Stable Ischemic Heart Disease

  • 22. 13 and 20 and 21

Baker WL. AHRQ CER Report on ACE Inhibitors and ARBs.

Current as of December 2009
Internet Citation: Searching for Relevant Studies (Text Version): Slide Presentation from the AHRQ 2009 Annual Conferenc. December 2009. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/news/events/conference/2009/white/index.html