SlideShare a Scribd company logo
1



Mini-manual for Clinical Questions & Ovid Medline Searching
Prepared by UMKC HSL Clinical Medical Librarians Susan Sanders and Kristy Steigerwalt


Part I - Clinical Questions & PICO Format

Introduction:
Every patient encounter leads to you asking clinical questions about the patient’s diagnosis,
treatment, management, or prognosis. These questions are similar to thinking about a
research project. They require a well-structured, logical question (hypothesis) followed by
a search for information (research), an explanation (discussion) and subsequently, a
decision about what action to take (conclusion). It’s no wonder that clinicians need
information systems to answer questions quickly and accurately. With the deluge of
information, it would be unreasonable to do a database search each time you had a
question, or when seeing a patient. There are systems such as UpToDate, DynaMed, ACP
PIER, and others that help you find information and answers rapidly, but there are good
reasons to learn how to search a biomedical databases such as Medline, CINAHL, or
PsychInfo, to name a few. What are your database search skills at present? What do you
need to learn about clinical questioning and literature searching?

A well-built clinical question usually has four parts, Patient, Intervention, Comparison,
Outcome – known as the PICO format. Here are three case examples to work with on
developing a PICO question.

Case Presentation #1:
The patient is a previously healthy, 81-year old woman with a history of progressive
dyspnea, leg edema and pallor over the last 3 months. She has no chronic conditions, is on
no medications, and requires no assistance. Her exam shows pallor, pitting edema of both
shins, diminished position and vibratory sensation over her feet and ankles, along with the
absence of neck vein distension or S3 gallop. Her chest x-ray is normal. Her test results
show very low hemoglobin, an increased MCV (mean corpuscular volume), and a very low
blood level of vitamin B12. What questions would you ask about this patient?

       Patient, Population or Problem
           o What are the characteristics of the patient or population?
           o What is the condition or disease you are interested in?
       Intervention or exposure
           o What do you want to do with this patient (e.g. treat, diagnose, observe)?
       Comparison
           o What is the alternative to the intervention (e.g. placebo, different drug,
               surgery)?
       Outcome
           o What are the relevant outcomes (e.g. morbidity, death, complications)?
2




Case Presentation #2:
Mr. Jones is a 55 year old man. He has just been diagnosed with Type 2 Diabetes.
Although you regularly prescribe Metformin to help regulate blood glucose levels in new
diabetics, Mr. Jones wants to try using regular exercise and diet to improve his blood
sugar. How will you treat him?

Case Presentation #3:
The patient presents with a calcium blood level of 17. She is given normal saline and
furosemide. Her condition improves, but what was the cause of her problem? What would
you ask?

Part II – Searching the Medline Database to Answer a Clinical Question

MEDLINE is the world’s largest database of indexed journal citations for the health
sciences literature, produced by the National Library of Medicine and National Center for
Biotechnology Information in Bethesda, Maryland. Medline is available in numerous
databases and is a free database through PubMed. There is some free full-text, but not
all. Medline is available through Ovid, Scopus, EBSCO, and other publishers with their own
unique interfaces, or “dashboards.”

Accessing Medline from UMKC’s Library Homepage
To Access Medline from UMKC, go to the library homepage
(http://library.umkc.edu/hslhome), use the dropdown menu and choose “HSL Quick Links”,
then select ”Medline Through Ovid”.
3




There are many ways to do searches on search engines and in databases. Databases
like Medline use “controlled vocabulary,” known as MeSH (Medical Subject Headings)
terms, Subheadings, and Text Words.

What is MeSH? http://www.nlm.nih.gov/mesh/

   1. MeSH is a standardized vocabulary for looking up concepts and terms, and referred
      to as the MeSH Tree. There are 16 main categories, or descriptors. Think of
      MeSH like an index in the back of a book where “terms” connect you to “pages.”
   2. In reality, there are indexers who really read new articles, and assign MeSH terms.
      They translate the authors’ non-uniform language into standardized MeSH
      Headings.
   3. To put MeSH to its best use, choose the single narrowest subject heading that
      corresponds with the information that you are trying to find.

Examples:
Different Terms, Same MeSH
Title #1, Treatment of gastric cancer.
Title #2, Technical considerations in laparoscopic resection of gastric neoplasms.
MeSH Heading for both titles: Stomach Neoplasms.

Same Term, Different MeSH
Title #1, The diagnosis of plaque-induced periodontal diseases.
MeSH Heading: Dental Plaque

Title #2, Mechanism of senile plaque formation in Alzheimer disease.
MeSH Heading: Senile Plaques

   4. Indexers also assign various sub-headings to MeSH Headings.

   Subheading – Secondary topic linked to a MeSH term by an indexer.

   EXAMPLE: *aortic rupture/su [Surgery] where the subheading is surgery

      Use a subheading only if it exactly matches a critical topic of your question.
      No subheading chosen? Then the system searches all of them that the indexer
       applied.

   EXAMPLE:
   Polycystic ovary syndrome/ge [Genetics] = 700+ citations
   Polycystic ovary syndrome AND Genetics [MeSH Major Topic] = 20 citations
4



Strengths of MeSH:

Standardization (controlled vocabulary) - Searcher doesn’t have to think or type
variations, like aortic rupture, rupture of the aorta, aortic tear, tear of aorta when using
MeSH.

Focus/Major Topic – Find articles where indexer tagged MeSH a major topic; omit
citations where the topic is a minor point. Helps target results, especially in a search for
only 1 topic. When searching two or more topics combined, try without major topic.

Subheadings – Indexer can pre-link 2 topics, e.g. Hernia, hiatal/su [Surgery]. MeSH +
subheading gets better results than hiatal hernia AND surgery, which incorrectly
retrieves irrelevant citations, e.g., hiatal hernia caused by surgery for reflux disease.

Mapping – Mapping is a function of the computer software. It suggests, or matches, your
typed words to a MeSH heading. For best mapping, type each term separately and
combine your searches. If mapping fails to find relevant articles, or in addition to
mapping, search title words, then display MeSH to see what MeSH terms were assigned to
the best results. Ask yourself, “What MeSH did indexers use for this topic?” and go with
what they used.

Weaknesses of MeSH:

1. MeSH headings don’t exist for all topics.
2. Very recent citations don’t have MeSH yet – indexing takes time.
3. Indexers don’t assign MeSH terms to every word in an abstract.
4. New MeSH terms are not retroactive, e.g., Individualized Medicine (2010) will not yield
   any pre-2010 results.
5. It is possible to miss the best MeSH terms.

MeSH Tree & Explode and Focus in Ovid Medline.




Look at the MeSH tree above. If you explode a term, the database searches for your
requested subject heading and any more specific terms that are related to your subject
5



heading. To find out what, if any, narrower topics are associated with your subject heading
(so that you can decide whether or not you want to explode your heading) click on the
heading itself. You will see a new screen that allows you to view the narrower (and
broader) subject terms associated with your topic. This screen shows subject headings in
a "tree" structure. Scroll down to find your subject heading in the tree. The plus (+) sign
next to a topic means that the topic has narrower terms associated with it. You can view
narrower terms by clicking on the plus sign. The number of articles that you can expect to
retrieve for each subject heading is provided after each heading.

Example Explode:

 From the subject heading screen you can click directly on "hypertension" to view its tree -
related terms that are more general and more specific. If you choose not to explode
hypertension, then you will miss articles that are indexed to more specific forms of
hypertension, such as renal hypertension. Instead, you will only retrieve articles about
hypertension that do not mention these more specific forms of the disease.




Focus
The focus command tells the database to retrieve only those articles in which your subject
term is considered to be the primary focus of the article. This will eliminate articles in
which your topic is discussed only peripherally. You will retrieve fewer (but probably more
relevant) articles if you use this feature.

You can explode and focus at the same time. If you check both the explode and focus
boxes next to your subject heading, the database will look for all of the exploded terms
6



and retrieve articles in which any one of the terms has been designated as the main focus
of the article.

Which term will explode? Neutropenia or Bird Diseases?

Subheadings

MeSH indexers apply subheadings to help further refine your question. The term Angina
Pectoris is assigned these allowable subheadings: /di (diagnosis), / th (therapy), /ep
(epidemiology), and so on. Agina Pectoris would not be assigned the subheading for
therapeutic use, /tu. Any citation assigned the MeSH term Angina Pectoris could be
assigned any or multiple allowable MeSH subheadings, and you need to be careful when
using subheadings because they limit a search. Here’s an example of how a subheading
limited and restricted search results ineffectively.

The patient presents with syncope, and is taking terfenadine, a "non-sedating"
antihistamine. You search for evidence on the risk of syncope in patients taking
antihistamines, and perform this Ovid Medline search:




Line 1: searches for syncope limited to the subheading etiology (ET).
Line 2: searches for Terfenadine and the appropriate group of antihistamines.
Line 3: Boolean AND combines the problem (syncope) and your intervention/exposure
(antihistamines) and yields 6 citations, none of them relevant.
Line 4: backtracking, you run a search filter for etiology or cause
Line 5: now search the disease as a key word (mp).
Line 6: Boolean AND combination of the search filter, disease, and the
intervention/exposure yields 17 citations. Two of these are pertinent including a large
HMO based cohort study.
7



The problem with the first search was using the wrong subheading for syncope. In the
cohort study that was relevant, syncope carried the subheadings CI (chemically induced)
and EP (epidemiology).

A search of antihistamines/co (complications) would have found nothing because CO is not
an allowable subheading for antihistamines. To find the allowable subheadings use the
"tools" function in Ovid. The subheading used in this article for antihistamines was AE
(adverse effects).

To become familiar with the indexing take a look at the "complete reference" for articles
which are relevant to your search and you can see exactly which subheadings are used,
then use them yourself.

SEARCHING WITH TEXT WORDS
You can use words from titles, abstracts, authors’ words, names of studies, etc. if no
MeSH matches a topic. Text word searching is a useful supplement to MeSH.

Strengths of text word searching
   - Instantly indexed; up-to-date terminology; every title, abstract word searchable
       (even low frequency terms); get results even if you miss the best MeSH.

Challenge of text word searching
  - Easy to miss information because authors use different words for the same topic.
   -   Can’t use Explode, Focus, or Subheadings – those features apply only to MeSH

Exercise: See if you can answers this…based on what you have read this far, would
you say Yes or No to the following “Features” when searching with either MeSH or
Text Word? For example, is Standardization a MeSH or a Text Word feature?

          Feature              MeSH                            Text Word (Title, Abstract)
      Standardization          YES                             NO
        Subheadings
        Major Topic
          Explode
       Latest Terms
       Fast Indexing
Finds Low Frequency Words
8



Word variations in text word searching
   1.   Use a truncation character (*) to search different word endings.

           o   Place an asterisk at the point where the variation begins, e.g. esophag* - retrieves
               esophagus, esophageal, esophagitis.
           o   Type synonyms and other word variations in search box, connected with OR
                      E.G., Autops* or post-mortem* or postmortem* Typing “or” bypasses
                       mapping and performs a “Multi-purpose” (.mp) search (title, abstract,
                       etc.) – works well for text words.
           o   The asterisk substitutes for a blank space or any number of characters. Can you
               think of other instances where it would be useful to use truncation?

   2. Synonyms:
           o   vitamin C – ascorbic acid
           o   renal failure – kidney failure – renal insufficiency


   3. Different spacing and punctuation
           o   RU486 RU 486 RU-486


   4. British Spellings
           o   e.g., paediatric* oesophag*


   5. Break up phrases
           o   Instead of elder abuse, try abus* AND elder* to retrieve these
                      abuse of the elderly, elder neglect and abuse, partner abuse among the
                       elderly
           o   Hartmann* retrieves hartmann’s procedure, Hartmann resection, Hartmann’s
               operation

Get help with text word variations
Search MeSH as a major topic then scan results to see word variations in titles.
1. Display the MeSH Scope Note
2. In Advanced Ovid Medline, Click [i] in Mapping or Tree Display - look at “Used For”

Take away: Add text words as safety net for MeSH. When results are important, use
both MeSH and text words.

Example MeSH +Text word search
The MeSH term: “Ebola Vaccines” was established in 2005-add the text word searches
ebola AND vaccin* to get pre 2005 results.
9



Part III: Putting It Together

1. Search each concept/idea/topic separately (most important first).
2. Create separate results sets for MeSH and text words.
3. Use MeSH mapping; check the Tree display to decide whether to explode; make major
   topic (focus) decision; check Subheadings for exact match to search topic. Use
   subheadings minimally.
4. Use Text Words – type synonyms, use truncation character, different spellings,
   spacings, etc.
5. Be sure topics in the text word search match topics in MeSH search.
6. If MeSH is Major topic, search text words in title only (major point will be found
   there).
7. Use parentheses to avoid logic errors when using AND/OR together, e.g.,
   Ovid: 1 AND (5 OR 8)
8. Use Ovid’s key operators & dot commands.

                                  Key Operators in Ovid
             *                Find alternate endings to         nurs* will find nurse,
                              this word                         nursing, nurses
            .tw.              Search for this term in the       anxiety.tw.
                              Title and Abstract fields
            Adj               Search for one term within        patient adj3 will find patient
                              x number of terms from            within three words of
                              another                           anxiety
           AND                Find articles where both          smoking AND cessation
                              terms appear
            OR                Find articles where either        smoking OR tobacco
                              term appears


9. If Ovid “Cannot map,” temporarily un-check “Map to MeSH” box – OR – type a field
   qualifier after the word in the search box: .mp, .ti These are also called Ovid DOT
   commands, and more are available in the table (next page).
10



                                    Ovid DOT Commands 1




10. Full text: First, try “Ovid Full Text” link, if available. If not, click Find it @ UMKC for
   UMKC full text. Order free through inter-library loan if not available from UMKC in
   either electronic or print.


Last Example! Question: Does dehydration cause delirium at the end of life?

More Related Content

PPT
Tutorial PUBMED
PPTX
PubMed Essentials
PPT
How to Search in PubMed® Tutorial
PPT
NLM for Health Sciences Student Session 3 - Using PubMed Pt. 2
PPT
PubMed: Searching and Search Features
PPT
Searching Pubmed and Medline
PPTX
Advanced PubMed Presentation (with Endnote)
PPT
Evidence-Based Health Care: A Tutorial Part 2
Tutorial PUBMED
PubMed Essentials
How to Search in PubMed® Tutorial
NLM for Health Sciences Student Session 3 - Using PubMed Pt. 2
PubMed: Searching and Search Features
Searching Pubmed and Medline
Advanced PubMed Presentation (with Endnote)
Evidence-Based Health Care: A Tutorial Part 2

What's hot (20)

PPTX
Lesson3#Scientific Research Skills
PPTX
Lit searchlsd
PPTX
PubMed: Basic Searching, Limits, and Scientific Journals
PPT
Searching Medline Using Dialog
PPTX
PPT
Ils Databases Powerpoint
PDF
Literature review
PPT
HDAS search guide
PPT
Outcomes B
PPT
Systematic Searching
PPT
Cancer nursing
PPTX
Finding scholarly nursing articles in databases
PPTX
Medline Resource Guide
PDF
CINAHL Effective Searching
PPT
Advanced Medline for Health Researchers
PPTX
1. Starting in Onesearch - gender selection
PPTX
2. Medline database searching sex preselection
PPTX
What's the difference between PubMed and Medline?
PDF
An Introduction to PubMed and MeSH
Lesson3#Scientific Research Skills
Lit searchlsd
PubMed: Basic Searching, Limits, and Scientific Journals
Searching Medline Using Dialog
Ils Databases Powerpoint
Literature review
HDAS search guide
Outcomes B
Systematic Searching
Cancer nursing
Finding scholarly nursing articles in databases
Medline Resource Guide
CINAHL Effective Searching
Advanced Medline for Health Researchers
1. Starting in Onesearch - gender selection
2. Medline database searching sex preselection
What's the difference between PubMed and Medline?
An Introduction to PubMed and MeSH
Ad

Similar to Mini manual-database-instruction-sanders (20)

DOCX
Original Work, NO PLAGERIESM, Cite Reference, References, 100 word.docx
PPT
How to Conduct a Systematic Search
PPT
RSS 2012 Literature Searches
PDF
Baylor Health Sciences Library PubMed Tutorial
PPT
Searching Databases to find Journal Articles Exercise Physiology 2014
PPTX
Medline Database searching
PDF
Nursing Resources
PDF
Nursing 1 p10 professional and therapeutic communications fall 2010
PDF
Searching the Literature: Search Techniques and Construction
PPT
How to Conduct a Literature Review
PPT
Advanced searching in pub med mar 2010
PPT
Advanced searching in pub med mar 2010
PPT
CINAHL Advanced Search Techniques
PPT
Cancer nursing
PPT
Systematic Reviews: Context & Methodology for Librarians
PPTX
PPTX
June 1st Library Presentation for CCTS Summer Fellowship
PPT
search-strategy.ppt
PPT
search-strategy.ppt
Original Work, NO PLAGERIESM, Cite Reference, References, 100 word.docx
How to Conduct a Systematic Search
RSS 2012 Literature Searches
Baylor Health Sciences Library PubMed Tutorial
Searching Databases to find Journal Articles Exercise Physiology 2014
Medline Database searching
Nursing Resources
Nursing 1 p10 professional and therapeutic communications fall 2010
Searching the Literature: Search Techniques and Construction
How to Conduct a Literature Review
Advanced searching in pub med mar 2010
Advanced searching in pub med mar 2010
CINAHL Advanced Search Techniques
Cancer nursing
Systematic Reviews: Context & Methodology for Librarians
June 1st Library Presentation for CCTS Summer Fellowship
search-strategy.ppt
search-strategy.ppt
Ad

Recently uploaded (20)

PDF
O7-L3 Supply Chain Operations - ICLT Program
PDF
Physiotherapy_for_Respiratory_and_Cardiac_Problems WEBBER.pdf
PDF
Sports Quiz easy sports quiz sports quiz
PPTX
Cell Types and Its function , kingdom of life
PDF
Supply Chain Operations Speaking Notes -ICLT Program
PPTX
GDM (1) (1).pptx small presentation for students
PDF
Pre independence Education in Inndia.pdf
PDF
01-Introduction-to-Information-Management.pdf
PDF
Computing-Curriculum for Schools in Ghana
PPTX
PPT- ENG7_QUARTER1_LESSON1_WEEK1. IMAGERY -DESCRIPTIONS pptx.pptx
PDF
Insiders guide to clinical Medicine.pdf
PPTX
Microbial diseases, their pathogenesis and prophylaxis
PDF
Anesthesia in Laparoscopic Surgery in India
PDF
Saundersa Comprehensive Review for the NCLEX-RN Examination.pdf
PPTX
IMMUNITY IMMUNITY refers to protection against infection, and the immune syst...
PPTX
Final Presentation General Medicine 03-08-2024.pptx
PPTX
Introduction_to_Human_Anatomy_and_Physiology_for_B.Pharm.pptx
PPTX
master seminar digital applications in india
PPTX
Pharma ospi slides which help in ospi learning
PDF
Black Hat USA 2025 - Micro ICS Summit - ICS/OT Threat Landscape
O7-L3 Supply Chain Operations - ICLT Program
Physiotherapy_for_Respiratory_and_Cardiac_Problems WEBBER.pdf
Sports Quiz easy sports quiz sports quiz
Cell Types and Its function , kingdom of life
Supply Chain Operations Speaking Notes -ICLT Program
GDM (1) (1).pptx small presentation for students
Pre independence Education in Inndia.pdf
01-Introduction-to-Information-Management.pdf
Computing-Curriculum for Schools in Ghana
PPT- ENG7_QUARTER1_LESSON1_WEEK1. IMAGERY -DESCRIPTIONS pptx.pptx
Insiders guide to clinical Medicine.pdf
Microbial diseases, their pathogenesis and prophylaxis
Anesthesia in Laparoscopic Surgery in India
Saundersa Comprehensive Review for the NCLEX-RN Examination.pdf
IMMUNITY IMMUNITY refers to protection against infection, and the immune syst...
Final Presentation General Medicine 03-08-2024.pptx
Introduction_to_Human_Anatomy_and_Physiology_for_B.Pharm.pptx
master seminar digital applications in india
Pharma ospi slides which help in ospi learning
Black Hat USA 2025 - Micro ICS Summit - ICS/OT Threat Landscape

Mini manual-database-instruction-sanders

  • 1. 1 Mini-manual for Clinical Questions & Ovid Medline Searching Prepared by UMKC HSL Clinical Medical Librarians Susan Sanders and Kristy Steigerwalt Part I - Clinical Questions & PICO Format Introduction: Every patient encounter leads to you asking clinical questions about the patient’s diagnosis, treatment, management, or prognosis. These questions are similar to thinking about a research project. They require a well-structured, logical question (hypothesis) followed by a search for information (research), an explanation (discussion) and subsequently, a decision about what action to take (conclusion). It’s no wonder that clinicians need information systems to answer questions quickly and accurately. With the deluge of information, it would be unreasonable to do a database search each time you had a question, or when seeing a patient. There are systems such as UpToDate, DynaMed, ACP PIER, and others that help you find information and answers rapidly, but there are good reasons to learn how to search a biomedical databases such as Medline, CINAHL, or PsychInfo, to name a few. What are your database search skills at present? What do you need to learn about clinical questioning and literature searching? A well-built clinical question usually has four parts, Patient, Intervention, Comparison, Outcome – known as the PICO format. Here are three case examples to work with on developing a PICO question. Case Presentation #1: The patient is a previously healthy, 81-year old woman with a history of progressive dyspnea, leg edema and pallor over the last 3 months. She has no chronic conditions, is on no medications, and requires no assistance. Her exam shows pallor, pitting edema of both shins, diminished position and vibratory sensation over her feet and ankles, along with the absence of neck vein distension or S3 gallop. Her chest x-ray is normal. Her test results show very low hemoglobin, an increased MCV (mean corpuscular volume), and a very low blood level of vitamin B12. What questions would you ask about this patient?  Patient, Population or Problem o What are the characteristics of the patient or population? o What is the condition or disease you are interested in?  Intervention or exposure o What do you want to do with this patient (e.g. treat, diagnose, observe)?  Comparison o What is the alternative to the intervention (e.g. placebo, different drug, surgery)?  Outcome o What are the relevant outcomes (e.g. morbidity, death, complications)?
  • 2. 2 Case Presentation #2: Mr. Jones is a 55 year old man. He has just been diagnosed with Type 2 Diabetes. Although you regularly prescribe Metformin to help regulate blood glucose levels in new diabetics, Mr. Jones wants to try using regular exercise and diet to improve his blood sugar. How will you treat him? Case Presentation #3: The patient presents with a calcium blood level of 17. She is given normal saline and furosemide. Her condition improves, but what was the cause of her problem? What would you ask? Part II – Searching the Medline Database to Answer a Clinical Question MEDLINE is the world’s largest database of indexed journal citations for the health sciences literature, produced by the National Library of Medicine and National Center for Biotechnology Information in Bethesda, Maryland. Medline is available in numerous databases and is a free database through PubMed. There is some free full-text, but not all. Medline is available through Ovid, Scopus, EBSCO, and other publishers with their own unique interfaces, or “dashboards.” Accessing Medline from UMKC’s Library Homepage To Access Medline from UMKC, go to the library homepage (http://library.umkc.edu/hslhome), use the dropdown menu and choose “HSL Quick Links”, then select ”Medline Through Ovid”.
  • 3. 3 There are many ways to do searches on search engines and in databases. Databases like Medline use “controlled vocabulary,” known as MeSH (Medical Subject Headings) terms, Subheadings, and Text Words. What is MeSH? http://www.nlm.nih.gov/mesh/ 1. MeSH is a standardized vocabulary for looking up concepts and terms, and referred to as the MeSH Tree. There are 16 main categories, or descriptors. Think of MeSH like an index in the back of a book where “terms” connect you to “pages.” 2. In reality, there are indexers who really read new articles, and assign MeSH terms. They translate the authors’ non-uniform language into standardized MeSH Headings. 3. To put MeSH to its best use, choose the single narrowest subject heading that corresponds with the information that you are trying to find. Examples: Different Terms, Same MeSH Title #1, Treatment of gastric cancer. Title #2, Technical considerations in laparoscopic resection of gastric neoplasms. MeSH Heading for both titles: Stomach Neoplasms. Same Term, Different MeSH Title #1, The diagnosis of plaque-induced periodontal diseases. MeSH Heading: Dental Plaque Title #2, Mechanism of senile plaque formation in Alzheimer disease. MeSH Heading: Senile Plaques 4. Indexers also assign various sub-headings to MeSH Headings. Subheading – Secondary topic linked to a MeSH term by an indexer. EXAMPLE: *aortic rupture/su [Surgery] where the subheading is surgery  Use a subheading only if it exactly matches a critical topic of your question.  No subheading chosen? Then the system searches all of them that the indexer applied. EXAMPLE: Polycystic ovary syndrome/ge [Genetics] = 700+ citations Polycystic ovary syndrome AND Genetics [MeSH Major Topic] = 20 citations
  • 4. 4 Strengths of MeSH: Standardization (controlled vocabulary) - Searcher doesn’t have to think or type variations, like aortic rupture, rupture of the aorta, aortic tear, tear of aorta when using MeSH. Focus/Major Topic – Find articles where indexer tagged MeSH a major topic; omit citations where the topic is a minor point. Helps target results, especially in a search for only 1 topic. When searching two or more topics combined, try without major topic. Subheadings – Indexer can pre-link 2 topics, e.g. Hernia, hiatal/su [Surgery]. MeSH + subheading gets better results than hiatal hernia AND surgery, which incorrectly retrieves irrelevant citations, e.g., hiatal hernia caused by surgery for reflux disease. Mapping – Mapping is a function of the computer software. It suggests, or matches, your typed words to a MeSH heading. For best mapping, type each term separately and combine your searches. If mapping fails to find relevant articles, or in addition to mapping, search title words, then display MeSH to see what MeSH terms were assigned to the best results. Ask yourself, “What MeSH did indexers use for this topic?” and go with what they used. Weaknesses of MeSH: 1. MeSH headings don’t exist for all topics. 2. Very recent citations don’t have MeSH yet – indexing takes time. 3. Indexers don’t assign MeSH terms to every word in an abstract. 4. New MeSH terms are not retroactive, e.g., Individualized Medicine (2010) will not yield any pre-2010 results. 5. It is possible to miss the best MeSH terms. MeSH Tree & Explode and Focus in Ovid Medline. Look at the MeSH tree above. If you explode a term, the database searches for your requested subject heading and any more specific terms that are related to your subject
  • 5. 5 heading. To find out what, if any, narrower topics are associated with your subject heading (so that you can decide whether or not you want to explode your heading) click on the heading itself. You will see a new screen that allows you to view the narrower (and broader) subject terms associated with your topic. This screen shows subject headings in a "tree" structure. Scroll down to find your subject heading in the tree. The plus (+) sign next to a topic means that the topic has narrower terms associated with it. You can view narrower terms by clicking on the plus sign. The number of articles that you can expect to retrieve for each subject heading is provided after each heading. Example Explode: From the subject heading screen you can click directly on "hypertension" to view its tree - related terms that are more general and more specific. If you choose not to explode hypertension, then you will miss articles that are indexed to more specific forms of hypertension, such as renal hypertension. Instead, you will only retrieve articles about hypertension that do not mention these more specific forms of the disease. Focus The focus command tells the database to retrieve only those articles in which your subject term is considered to be the primary focus of the article. This will eliminate articles in which your topic is discussed only peripherally. You will retrieve fewer (but probably more relevant) articles if you use this feature. You can explode and focus at the same time. If you check both the explode and focus boxes next to your subject heading, the database will look for all of the exploded terms
  • 6. 6 and retrieve articles in which any one of the terms has been designated as the main focus of the article. Which term will explode? Neutropenia or Bird Diseases? Subheadings MeSH indexers apply subheadings to help further refine your question. The term Angina Pectoris is assigned these allowable subheadings: /di (diagnosis), / th (therapy), /ep (epidemiology), and so on. Agina Pectoris would not be assigned the subheading for therapeutic use, /tu. Any citation assigned the MeSH term Angina Pectoris could be assigned any or multiple allowable MeSH subheadings, and you need to be careful when using subheadings because they limit a search. Here’s an example of how a subheading limited and restricted search results ineffectively. The patient presents with syncope, and is taking terfenadine, a "non-sedating" antihistamine. You search for evidence on the risk of syncope in patients taking antihistamines, and perform this Ovid Medline search: Line 1: searches for syncope limited to the subheading etiology (ET). Line 2: searches for Terfenadine and the appropriate group of antihistamines. Line 3: Boolean AND combines the problem (syncope) and your intervention/exposure (antihistamines) and yields 6 citations, none of them relevant. Line 4: backtracking, you run a search filter for etiology or cause Line 5: now search the disease as a key word (mp). Line 6: Boolean AND combination of the search filter, disease, and the intervention/exposure yields 17 citations. Two of these are pertinent including a large HMO based cohort study.
  • 7. 7 The problem with the first search was using the wrong subheading for syncope. In the cohort study that was relevant, syncope carried the subheadings CI (chemically induced) and EP (epidemiology). A search of antihistamines/co (complications) would have found nothing because CO is not an allowable subheading for antihistamines. To find the allowable subheadings use the "tools" function in Ovid. The subheading used in this article for antihistamines was AE (adverse effects). To become familiar with the indexing take a look at the "complete reference" for articles which are relevant to your search and you can see exactly which subheadings are used, then use them yourself. SEARCHING WITH TEXT WORDS You can use words from titles, abstracts, authors’ words, names of studies, etc. if no MeSH matches a topic. Text word searching is a useful supplement to MeSH. Strengths of text word searching - Instantly indexed; up-to-date terminology; every title, abstract word searchable (even low frequency terms); get results even if you miss the best MeSH. Challenge of text word searching - Easy to miss information because authors use different words for the same topic. - Can’t use Explode, Focus, or Subheadings – those features apply only to MeSH Exercise: See if you can answers this…based on what you have read this far, would you say Yes or No to the following “Features” when searching with either MeSH or Text Word? For example, is Standardization a MeSH or a Text Word feature? Feature MeSH Text Word (Title, Abstract) Standardization YES NO Subheadings Major Topic Explode Latest Terms Fast Indexing Finds Low Frequency Words
  • 8. 8 Word variations in text word searching 1. Use a truncation character (*) to search different word endings. o Place an asterisk at the point where the variation begins, e.g. esophag* - retrieves esophagus, esophageal, esophagitis. o Type synonyms and other word variations in search box, connected with OR  E.G., Autops* or post-mortem* or postmortem* Typing “or” bypasses mapping and performs a “Multi-purpose” (.mp) search (title, abstract, etc.) – works well for text words. o The asterisk substitutes for a blank space or any number of characters. Can you think of other instances where it would be useful to use truncation? 2. Synonyms: o vitamin C – ascorbic acid o renal failure – kidney failure – renal insufficiency 3. Different spacing and punctuation o RU486 RU 486 RU-486 4. British Spellings o e.g., paediatric* oesophag* 5. Break up phrases o Instead of elder abuse, try abus* AND elder* to retrieve these  abuse of the elderly, elder neglect and abuse, partner abuse among the elderly o Hartmann* retrieves hartmann’s procedure, Hartmann resection, Hartmann’s operation Get help with text word variations Search MeSH as a major topic then scan results to see word variations in titles. 1. Display the MeSH Scope Note 2. In Advanced Ovid Medline, Click [i] in Mapping or Tree Display - look at “Used For” Take away: Add text words as safety net for MeSH. When results are important, use both MeSH and text words. Example MeSH +Text word search The MeSH term: “Ebola Vaccines” was established in 2005-add the text word searches ebola AND vaccin* to get pre 2005 results.
  • 9. 9 Part III: Putting It Together 1. Search each concept/idea/topic separately (most important first). 2. Create separate results sets for MeSH and text words. 3. Use MeSH mapping; check the Tree display to decide whether to explode; make major topic (focus) decision; check Subheadings for exact match to search topic. Use subheadings minimally. 4. Use Text Words – type synonyms, use truncation character, different spellings, spacings, etc. 5. Be sure topics in the text word search match topics in MeSH search. 6. If MeSH is Major topic, search text words in title only (major point will be found there). 7. Use parentheses to avoid logic errors when using AND/OR together, e.g., Ovid: 1 AND (5 OR 8) 8. Use Ovid’s key operators & dot commands. Key Operators in Ovid * Find alternate endings to nurs* will find nurse, this word nursing, nurses .tw. Search for this term in the anxiety.tw. Title and Abstract fields Adj Search for one term within patient adj3 will find patient x number of terms from within three words of another anxiety AND Find articles where both smoking AND cessation terms appear OR Find articles where either smoking OR tobacco term appears 9. If Ovid “Cannot map,” temporarily un-check “Map to MeSH” box – OR – type a field qualifier after the word in the search box: .mp, .ti These are also called Ovid DOT commands, and more are available in the table (next page).
  • 10. 10 Ovid DOT Commands 1 10. Full text: First, try “Ovid Full Text” link, if available. If not, click Find it @ UMKC for UMKC full text. Order free through inter-library loan if not available from UMKC in either electronic or print. Last Example! Question: Does dehydration cause delirium at the end of life?