Chapter 3 Lecture Research Techniques: For the Health Sciences Fifth Edition © 2014 Pearson Education, Inc. Critical Review of the Literature and Information

Chapter 3 Lecture

© 2014 Pearson Education, Inc.

Research Techniques: For the Health Sciences

Fifth Edition

Critical Review of the Literature and Information Sources

R. Eric Heidel, PhD

University of Tennessee Graduate School of Medicine


Review of the Literature

• Relevant Literature– provides a framework for the research

question, hypothesis, and setting of the problem


Purposes of the Review

• The general purpose is to gain an understanding of previous work and to generate new ideas and concepts.

• The process can additionally help the investigator to:

1. develop an understanding and grounding in theory

2. define the problem

3. review the procedures and instruments used

4. originate new ideas rather than repeat work already accomplished

5. use the recommendations for further research

6. critique relevant studies


The Importance of the Research Question

• The key words used in any literature search are the basis for a successful review of the literature.

• Clinically based questions demand timely answers and the best available evidence.

• The best evidence requires the most answerable clinical questions.


The Importance of the Research Question (cont'd)• The PICO acronym is used as a framework for

developing research questions and for conducting literature searches– P—patient or problem– I—intervention– C—comparison– O—outcome


Steps in the Review Process

• Steps to consider when beginning a review of the literature:1. reading background information2. gathering the necessary tools3. listing key words4. checking preliminary sources, including

databases5. conducting a computer search6. determining what to read7. determining the level of evidence8. finding shortcuts to determining the level of

evidence


Reading Background Information

• Primary Sources– written by someone who has observed or

participated in an event– must be read– include journals, final reports, or books that

contain original research– government publications are good primary

sources for many health topics


Reading Background Information (cont'd)

• Secondary Sources– textbooks written by someone who has not

directly observed the described event– usually have a bibliography, which provides

the reader with relevant primary sources


Gathering the Necessary Tools

• Systematically gathering data will keep you organized and will prevent you from having to redo what you have already done.

• Bibliography Index– Filing systems may be arranged by:

1. authors' names, in alphabetical order

2. date, with the most recent work first

3. subheading

4. level of evidence

– A combination of these techniques could also be used


Listing Key Words

• After collecting the background information, you will have an idea of the topic area and be able to generate key words.

• Key words or phrases are necessary because almost all health science sources are organized by subject, and you should have a list of key words to search the computer databases.


Checking Preliminary Sources

• Preliminary sources– help the investigator locate primary sources

• reference books• indices• abstracts• guidebooks• periodicals


General Indexing and Abstracting Services

• There are several databases available through online computer services at most libraries.

• A growing number of these databases contain full text or direct you to websites with full text of the document.


General Indexing and Abstracting Services (cont'd)• Databases:

– BIOETHICSLINE– Biological Abstracts– CINAHL (Cumulative

Index to Nursing and Allied Health Literature)

– ERIC– HAPI (Health and

Psychosocial Instruments)

– MEDLINE– MEDLINEplus– National Library of

Medicine– PsycINFO– Social Work

Abstracts– TOXNET


Evidence-Based Full-Text and Abstracting Services• These database services can save the health researcher

an immense amount of time when seeking high-quality evidence-based information.

• Examples:– Bandolier– Cochrane Library– Database of Abstracts of Reviews of Effectiveness

(DARE)– National Guideline Clearinghouse (NGC)– PedsCCM Evidence-Based Journal Club– PubMed– SUMSearch– TRIPdatabase (Turning Research into Practice)


Government Documents

• These listings can be accessed as government documents using the website of the Centers for Disease Control for the National Center for Health Statistics (www.cdc.gov/nchs)– National Center for Health Statistics (NCHS)– National Health and Nutrition Examination Survey

(NHANES)– National Health Care Survey (NHCS)– National Health Interview Survey (NHIS)– National Immunization Survey (NIS)– National Survey of Family Growth (NSFG)– National Vital Statistics System (NVSS)


Conducting a Computer Search: Finding the Evidence• Using the computer search will enhance your

ability to check the preliminary sources.• Full-text electronic journals and abstracts are

available using a literature search engine.• Some online resources are available strictly

through university libraries and public libraries.• The printouts of the list of citations will include a

full bibliographic entry that can later be converted to a bibliographic database of your own.


How to Conduct a Computer Search

1. Specify the research problem.

2. Select the databases.

3. Select the descriptors.

4. Conduct the search.

5. Increase sensitivity and specificity.

6. Review the citation list.


Determining the Level of Evidence

• As a general rule, useful information must have three attributes:

1. It must be relevant to the research effort.

2. It must be correct.

3. It must require little effort to procure.• The formula is:

level of evidence =(relevance x validity)

work


Determining the Level of Evidence (cont'd)

• Relevance– This component begins with the applicability

of the evidence to your problem.– Information must be critically appraised or

evaluated for its validity and research usefulness.



• Validity– Health literature can be grouped into a

pyramid of evidence• Categories range single studies that provide little

evidence to systematic reviews that collect and describe all relevant studies addressing the research question.

• Researchers, among others, want to base their decisions on the best evidence available.


Haynes's 5 Ss



• Work– In health care, there are several databases

that allow the researcher to search by level of evidence.

– Databases that conduct systematic reviews are preferred because they locate, appraise, and synthesize evidence from scientific studies in order to provide informative empirical answers to scientific research questions.

– Systematic reviews adhere to a strict scientific design in order to be more comprehensive, minimize bias, and ensure reliability.


Determining the Quality of Evidence

• The GRADE system rates the quality of evidence and grades the strength of recommendations for systematic reviews, health technology assessments, and clinical practice guidelines looking at alternative management options.

• The system classifies the quality of evidence as high, moderate, low, or very low according to factors that include the study methodology, consistency, and precision of the results and directness of the evidence.

• A key in using GRADE is a well-worded clinical question with possible outcomes.


Writing the Section on Related Literature

• Relating the Review– Information in the literature review should

always relate to the introductory material.– This will enable the introduction to flow

coherently and present an organized approach to theory and research related to your topic.

– Literature should be related to the purpose of the study, the generated hypotheses, and the population in question.


Writing the Section on Related Literature (cont'd)• Developing a Plan

– Subheadings are usually based on the variables and their relationship to the problem of your study.

– Begin each subtopic with an introductory sentence and end with a summarizing section.


Writing the Section on Related Literature (cont'd)• Summarizing

– This section recaps the relevant information relating to theory, previous research, new insights, and the stated hypotheses.

– Generally, one or two paragraphs should suffice, if presented cogently.