doi:10.1016/S0899-3467(07)60142-6

Writing narrative literature reviews for

peer-reviewed journals: secrets of

the trade

Bart N. Green, DC, MSEd, DACBSP

Claire D. Johnson, DC, MSEd, DACBSP

Alan Adams, DC, MS, MSEd, DACBN

Associate Editor, National University of Health Sciences.

Editor, National University of Health Sciences.

Vice President of Academic Affairs and Program Development, Texas Chiropractic College

Submit requests for reprints to: Dr. Bart Green, National University of Health Sciences, 200 E. Roosevelt Rd,

Lombard, IL 60148, [email protected]

Sources of support: This article is reprinted with permission. Its original citation is: Green BN, Johnson CD,

Adams A. Writing narrative literature reviews for peer-reviewed journals: secrets of the trade. J Sports

Chiropr Rehabil 2001;15:5–19.

ABSTRACT

Objective: To describe and discuss the process used to write a narrative review of the literature for publication in

a peer-reviewed journal. Publication of narrative overviews of the literature should be standardized to increase

their objectivity.

Background: In the past decade numerous changes in research methodology pertaining to reviews of the literature

have occurred. These changes necessitate authors of review articles to be familiar with current standards in the

publication process.

Methods: Narrative overview of the literature synthesizing the findings of literature retrieved from searches of

computerized databases, hand searches, and authoritative texts.

Discussion: An overview of the use of three types of reviews of the literature is presented. Step by step instructions

for how to conduct and write a narrative overview utilizing a ‘best-evidence synthesis’ approach are discussed,

starting with appropriate preparatory work and ending with how to create proper illustrations. Several resources

for creating reviews of the literature are presented and a narrative overview critical appraisal worksheet is

included. A bibliography of other useful reading is presented in an appendix.

Conclusion: Narrative overviews can be a valuable contribution to the literature if prepared properly. New and

experienced authors wishing to write a narrative overview should find this article useful in constructing such a

paper and carrying out the research process. It is hoped that this article will stimulate scholarly dialog amongst

colleagues about this research design and other complex literature review methods. (J Chiropr Med 2006;5:101–

117)

Key Indexing Terms: Review Literature; Authorship; Peer Review, research; Manuscripts; Meta-

analysis

Clinical Update

101

0899-3467/Clinical Update/1002-049$3.00/0

JOURNAL OF CHIROPRACTIC MEDICINE

INTRODUCTION

Background

The purpose of this article is to describe and discuss

the research design known as a review of the litera-

ture and to delineate how to write a particular vari-

ety of this research design, the narrative overview of

the literature. Another intention of this article is to

provide educational information and assistance for

those who have not yet published a literature re-

view and to decrease potential author frustration

that can arise during the peer review process. It is

important to note that the general classification of

’literature review’ has three varieties: narrative re-

view, qualitative systematic review and quantitative

systematic review. Each will be addressed in this

article. However, the primary focus of this article

will be on the writing of a narrative review.

A literature review is a type of research article pub-

lished in a professional peer-reviewed journal. The

purpose of a literature review is to objectively report

the current knowledge on a topic and base this

summary on previously published research.

A lit-

erature review provides the reader with a compre-

hensive overview and helps place that information

into perspective.

The literature review research design is different

from other research designs because rather than

patients, data to write the report are collected from

the published literature.

3,4

These full length articles

provide a new conclusion to the literature, not the

brief summary of literature that is given typically in

the introduction or discussion sections of other re-

search designs.

2,5

In creating a literature review, the

author searches through the literature, retrieves nu-

merous sources of information and synthesizes the

findings of all relevant sources into one article.

2,3,6

Thus, a vast amount of information is brought to-

gether and written in a manner in which the reader

can clearly understand the topic.

There are several reasons to read reviews of the

literature. For the clinician, they can save valuable

time when reviewing or searching for information

about patient care by condensing a great amount of

information into a few pages.

1,6,7

The clinician can

read one paper instead of sifting through the whole

of the literature to find the answer to a clinical

question; the author of the literature review has

already done most of this work for him. Literature

reviews also provide information for decision mak-

ers and are used by researchers to identify, justify

and refine hypotheses and to recognize and avoid

pitfalls in previous research.

1,8

Additionally, reviews

of the literature provide a basis for validating as-

sumptions,

provide insight into the dynamics un-

derlying the findings of other studies

and may

offer more conclusive results than a single primary

research study.

Depending on the variety of litera-

ture review, they may provide a very high level of

evidence for making clinical practice decisions.

One of the cautions that one must consider with

literature reviews is the bias that is often associated

with them.

10,11,12,13

As an author, it is important to

attempt to reduce bias as much as possible through

appropriate writing and research techniques. An in-

crease in objectivity leads to improved utility and

credibility in publications.

While certain criteria

for literature reviews have been published,

little

has been accomplished in terms of standardizing

and verifying the validity of the criteria pro-

posed.

1,14

Indeed, many changes have taken place

in recent years regarding publication standards for

literature reviews

10,12

and it is important for authors

to keep current with these changes. This paper

clearly states the minimum acceptable criteria they

pertain to narrative overviews of the literature.

METHODS

Information used to write this paper was collected

from the sources listed in table 1.

DISCUSSION

Three Varieties of Reviews of the Literature

The three basic types of literature reviews are nar-

rative reviews, qualitative systematic reviews, and

Table 1

Sources Used for This Overview

• MEDLINE search 1966–January 2001. Key words:

Review of the Literature; Authorship; Meta-analysis;

Narrative overview.

• CINAHL search from 1982 to December 2000. Key

words: Review of the Literature; Authorship;

Meta-analysis; Narrative overview.

• Hand searches of the references of retrieved

literature.

• Personal and college libraries searching for texts

on research methods and literature reviews.

• Discussions with experts in the field of reviews of

the literature.

• Personal experience participating in and writing

several reviews of the literature.

Volume 5 • Number 3 •

Fall 2006

102

quantitative systematic reviews (meta-analyses).

The amount of clinical evidence afforded by each of

these designs increases as the methods employed to

conduct them become more detailed and elaborate.

In this section, the emphasis will be placed on nar-

rative reviews, since they are the subject of this

article; a brief description of qualitative and quanti-

tative systematic reviews will also be given.

Narrative Literature Review

There are three types of narrative reviews of the

literature: editorials, commentaries, and overview

articles.

4,15

Editorials, typically written by the editor of the jour-

nal or an invited guest, may be a narrative review if

the author retrieves and synthesizes information

about a particular topic for the reader. Usually these

types of narrative reviews are based upon a short,

select and narrowly focused review of only a few

papers.

However, editorials may be no more than

the editor’s comments regarding a current issue of

the journal or a current event in health care. There-

fore, editorials do not automatically qualify as nar-

rative reviews.

Commentaries may also be written as a narrative

review, however they are typically written with a

particular opinion being expressed.

In these articles

research methodology is usually not presented and

the author’s synthesis of the articles demonstrates

bias. Commentaries are usually shorter than a full

length review article and it is expected that the

author possesses expertise in the content area of the

commentary. In short, a commentary is a biased

narrative review that draws upon the wisdom of the

commentator. Usually the purpose of a commentary

is to provoke scholarly dialog among the readers of

the journal.

Narrative overviews, also known as an unsystematic

narrative reviews,

are comprehensive narrative syn-

theses of previously published information. The de-

tails of how to prepare this type of article are pre-

sented in this paper. This type of literature review

reports the author’s findings in a condensed format

that typically summarizes the contents of each ar-

ticle.

Some researchers suggest that a proper narra-

tive overview should critique each study in-

cluded,

2,17

but other authors write that this is not

necessarily a property of overviews.

It is up to the

author to determine which of these two paths to

take when writing the article.

There are many good reasons to write a quality

narrative overview. Narrative overviews are useful

educational articles since they pull many pieces of

information together into a readable format. They

are helpful in presenting a broad perspective on a

topic and often describe the history or development

of a problem or its management.

2,10

Faculty like to

use overviews in the classroom because they are

often more up to date than textbooks, provide a

single source for students to read from, and expose

students to peer reviewed literature. Narrative over-

views are also used as educational articles to bring

practitioners up to date with certain clinical proto-

cols.

7,11

Some journals, publish quizzes related to

such articles; these quizzes can be submitted to

regulating boards for continuing education credit.

Often discussing theory and context, narrative over-

views can serve to provoke thought and contro-

versy. For this reason, these reviews may be an

excellent venue for presenting philosophical per-

spectives in a balanced manner. Philosophical ar-

ticles can be excellent for stimulating scholarly dia-

log amongst readers. Readers can participate in this

process by writing to the letters to the editor section

of the journal and present their opinions and critical

appraisal. The letters to the editor section can be a

dynamic part of a journal; several times in the his-

tory of health care tremendous insight into patient

management and research design has been pre-

sented in this forum.

Authors of narrative overviews are often acknowl-

edged experts in the field and have conducted re-

search themselves.

6,7,15,18

Editors sometimes solicit

narrative overviews from specific authors in order to

bring certain issues to light.

Authors must be care-

ful to avoid a common pitfall of the overview de-

sign, which is to present an opinion oriented argu-

ment based upon a myriad of references,

rather

than objective conclusions based upon the literature

reviewed. For this reason, some studies have deter-

mined that some experts are less likely to adhere to

high levels of methodological rigor when writing

these papers than non-experts.

Therefore, wheth-

er one is a novice or expert, the critical factor in

writing a good narrative review is to use good meth-

ods.

Once quite common, overviews are slowly falling

into disfavor in some journals due to a lack of

Fall 2006

• Number 3 • Volume 5

103

systematic methods that should be employed to

construct them.

11,19

Rarely have the methods used

in creating the paper been divulged to the

reader,

1,10,11

which is a problem identified as early

as 1987.

Usually the number of sources employed

to find the literature are incomplete,

4,10

possibly

creating an insignificant knowledge base from

which to draw a conclusion. In this rather unsys-

tematic approach, selection of information from

primary articles is usually subjective, lacks explicit

criteria for inclusion and leads to a biased re-

view.

4,7,10,13,16

The author’s interpretation and syn-

thesis of information should take into account ma-

jor differences between studies, such as if patients

samples in one study are completely different than

in another

or that research designs are not compa-

rable.

3,7

Without identifying these differences, one

takes the risk of providing faulty conclusions or

incorrect information. All of these potential pitfalls

are avoidable if the author is aware of them and

takes the appropriate steps to avoid them.

In the past, many reviews of the literature were

constructed based upon the personal papers of the

author, creating a bias that was slanted to what that

author found interesting or controversial.

When

this occurs it is difficult to discern if the author has

constructed an objective review of the literature or a

lengthy commentary. Biased writers will draw con-

clusions based more on opinion than data, which is

not a truthful representation of the research.

13,16

Often times this faulty synthesis is then repeated by

other authors and the mistakes are handed down

from one study to the next,

thus perpetuating the

errors. The aforementioned problems related to lit-

erature reviews are a potential danger in health care

if readers make patient health care decisions based

upon faulty reviews.

13,18

While narrative overviews are great papers to read

to keep up to date, receive continuing education

credits, or challenge your way of thinking, they are

not a form of evidence that should be used fre-

quently when making decisions about how to solve

specific clinical patient problems. Narrative over-

views are one of the weakest forms of evidence to

use for making clinical decisions in regard to patient

care,

primarily because they deal more with

broader issues than focused clinical problems. Addi-

tionally, there is a higher degree of bias involved in

overviews than some other research designs.

21,22

Nevertheless, narrative overviews constitute an im-

portant component in the literature base.

Qualitative Systematic Literature Review

A systematic review is a type of literature review

that employs detailed, rigorous and explicit meth-

ods.

4,22,23

A detailed search of the literature based

upon a focused question or purpose is the hallmark

of a systematic review.

4,22

Since the review is struc-

tured around a focused clinical question, it allows

the researcher to develop criteria that determine if a

research publication should be included or excluded

in the final synthesis.

22,23

Step-by-step methodology

used in the research is described. Authors of system-

atic reviews attempt to obtain all original (primary)

research studies published on the topic under study

by searching in multiple databases, performing hand

searches and contacting authors of previously pub-

lished research. Additionally, authors will attempt

to locate articles that may not have been published

because the results of the study did not support the

research hypothesis.

1,4,22,23

Each paper is reviewed in a systematic and consis-

tent manner, usually by several independent re-

viewers, and usually rated using a scoring system by

the authors.

1,4,23

Each piece of evidence drawn from

a paper for the literature review is extracted in the

same fashion to help decrease the bias that occurs

when this information extraction is done subjec-

tively, such as in a narrative overview.

Authors

create data, or evidence tables, in order to tease out

the differences in the results of different studies.

These reviews of the literature are called qualitative

because the process by which the individual studies

are integrated includes a summary and critique of

the findings derived from systematic methods, but

does not statistically combine the results of all of the

studies reviewed.

22,24

Because of the rigorous methods employed in con-

ducting qualitative systematic reviews, they are a

more powerful evidence-based source to garner

clinical information than narrative reviews, case re-

ports, case series, and poorly conducted cohort stud-

ies.

Quantitative Systematic Literature

Review (Meta-analyses)

A systematic review that critically evaluates each

paper and statistically combines the results of the

studies is called a quantitative systematic review of

the literature, also known as a meta-analy-

sis.

1,10,22,23,24

Introduced in 1976, meta-analyses

Volume 5 • Number 3 •

Fall 2006

104

aim to make an objective science out of research

synthesis.

Meta-analyses employ all of the rigor-

ous methodology of qualitative systematic reviews.

In addition to the inherent strengths of the system-

atic review process, the major benefit of a meta-

analysis is the pooling of data between studies. In

this process, the authors of a meta-analysis will

gather the original patient data from each of the

studies under review, pool it all together in a data-

base, and perform the appropriate statistics on this

larger sample size. This is especially useful when

clinical trials exist in the literature but possess low

sample sizes that prevent the authors from making

conclusions that can be generalized to the popula-

tion at large.

This can be particularly powerful if

the studies under review are very similar in their

construction because several studies can be com-

bined as one larger base of data leading to more

powerful conclusions.

The pooling of data that can be analyzed statistically,

which is the strength of the meta-analysis,

can

also be a drawback because it is difficult to find

studies that are similar enough to one another to

draw valid comparisons.

4,25,26

There is disagreement

amongst experts about the most appropriate meth-

ods to combine the data from studies with different

variables (eg, patient populations, clinical outcome

measures, treatments).

25,26,27

Meta-analyses are considered to be a very high form

of evidence for making clinical decisions because the

results of the review are produced from a rigorous

critical appraisal and pooling of data from the stud-

ies reviewed.

11,12,16,26

This leads to a more general-

izable conclusion.

Writing a Narrative Review of the Literature,

Step By Step

Preparation

Before involving oneself completely in the research

endeavor, it is important to observe a few tips that

will sustain the author during the process of the

project. These pearls of wisdom, presented in table

2, are gathered from experienced authors and are

useful for novice as well as experienced writers.

The first step in writing a narrative overview is to

perform a preliminary search of the literature.

this endeavor, the author should search the litera-

ture to see what other work in the area of interest

has already been published.

5,17

This initial work

should help the author to refine the topic and ob-

jective of the overview being written. For example,

if one wishes to publish a review of the literature

about the effect of chiropractic adjustments on cer-

vical spine pain, the initial search should reveal if

someone has recently published such a study. In

addition, this initial work will also give the author a

preview of the number of articles available on the

topic. If an article is already published on the very

same topic that the author wishes to write about,

then it may be better to select a slightly different

topic or to slightly modify the focus of the objective.

On the other hand, if there has never been a review

published about the topic, then this helps to estab-

lish the need for this particular contribution.

5,17

Once armed with a refined topic, it is time to pro-

ceed with the next steps.

General guidelines

The text that follows delineates what would prob-

ably be acceptable in many journals. These recom-

Table 2

Words of Wisdom for Authors Writing Narrative Overviews

Interesting Topic Select a topic that you are very interested in, lest you do a lot of work and

then lose momentum to finish the project. There are many half-finished

papers collecting dust on shleves because authors lost the drive and

interest to complete the task. Select an enticing and engaging topic that will

keep you fascinated throughout the process.

Doable Project Select a topic with a feasible focus. A narrative review on ‘headaches’ is an

impossible task, whereas ‘chiropractic management of muscle tension

headaches’ can be a manageable narrative review. Keep the focus clear and

defined and you will be able to complete the paper.

Get Help Get help early and often. Call upon people around you who have writing

experience, such as colleagues or faculty who have published narvative

reviews. Consider asking them for their opinions before you begin the paper

and then ask them to review drafts of your paper before submitting it. Some

journal editors can direct you to others that may be of assistance and may be

able to help you themselves when time permits.

Fall 2006

• Number 3 • Volume 5

105

mendations follow the approach put forth by Robert

Slavin, which is called a “best-evidence synthesis”.

However, there are no rigid published guidelines

designating exactly what should or should not be

included in a quality narrative overview.

A suc-

cessful narrative review should do the following:

present information that is written using the re-

quired elements for a narrative review, be well

structured, synthesize the available evidence per-

taining to the topic, and convey a clear message.

An objective and scientific approach on behalf of the

author should be conveyed and the paper should

follow the formatting guidelines published in the

Uniform Requirements for Manuscripts Submitted to Bio-

medical Journals. These guidelines describe in detail

what is necessary in order to prepare a manuscript

for submission to a peer reviewed journal. Preparing

a manuscript using these guidelines is essential to

insure that manuscripts are uniform in nature, as

objective as possible, and can be processed by edi-

tors in an expeditious manner, thus providing the

author with a better chance for earlier accep-

tance.

28,29

The Uniform Requirements can be found

on the Internet at http://www.icmje.org/

index.html.

The presentation of a narrative review should be as

objective as possible. It is essential that prospective

authors remember that the intention of a narrative

review is to describe and synthesize the available

literature on a topic, providing a conclusion from

this evidence.

The necessary elements of a narra-

tive review are similar to those required of any form

of scholarly article (Table 3). A detailed rating scale

for narrative reviews of the literature is also in-

cluded in this article as Appendix A.

Title

The title of the article should be interesting and

clearly describe the topic being reviewed.

It is also

helpful to readers if the words “literature review” or

“review of the literature” appear somewhere in the

title to make it clear to readers what research design

is being used. The reader instantly knows what the

major topic is and that previous publications about

the topic are being synthesized. A good example of a

title is Injuries Associated with Soccer: A Review of Epi-

demiology and Etiology. A bad example of a literature

review title might be The Epidemiology of Soccer Inju-

ries. This title would infer that a population based

study was performed to determine the kinds and

frequency of soccer injuries, which would be mis-

leading to the reader.

Abstract

The abstract is a structured summary of the article

that offers the reader a brief presentation of the

review and relates the most important informa-

tion.

The abstract and the title are entered into

computer databases and indexing systems, and are

essentials for those conducting literature searches. A

well written abstract allows people searching the

literature to find the information in their search and

discern whether or not they should retrieve the

paper.

29,30

Abstracts should be written in a structured format.

Structured abstracts are required in order to assure

that all necessary information for an abstract is re-

ported for the reader. In the past, narrative abstracts

were often used by journals, but authors sometimes

did not adequately report the necessary elements of

the study in the abstract. Thus, most journals

adopted the structured abstract format over ten

years ago.

30,31

Subsections for abstracts of narrative

overviews of the literature typically consist of the

following, further described in table 4: objective,

background, methods, discussion; conclusion, key

words.

Introduction

Early in the introduction, the author should clearly

state the research purpose or focus. Clearly stated

aims tell the reader that the study was planned out

in advance, usually resulting in a well outlined

study that presents useful information.

3,18,24

Next, the author must make a case for the need or

importance of the study. This is essential in order to

relate the importance of this research to the reader.

This usually requires that the author has already

reviewed the literature pertaining to the topic and

Table 3

Components of a Narrative Overview in

Order of Appearance in the Manuscript

1. Title

2. Structured abstract

3. Introduction

4. Methods

5. Discussion

6. Conclusion

7. Acknowledgements (if applicable)

8. References

9. Tables

10. Figures

11. Figure captions

Volume 5 • Number 3 •

Fall 2006

106

has discovered a deficiency of a well written review

of the literature in the area. Authors can use the

results of their preparatory literature search dis-

cussed previously as well as further reading un-

veiled during the course of conducting the formal

literature review, described later, for this purpose.

This information should be written in the introduc-

tion to state why the study is important and place it

in context with other published information.

10,24

The author should also define any unusual terms or

words that are essential to understanding the infor-

mation in the paper.

For example, if the paper is

about isometric low back extension endurance tests,

it would be paramount that the author clearly de-

fine what these tests are and what they do.

Methods

The methods section of the article should describe

step by step how the study was performed.

11,14,18,24

Sources of information: The most efficient way to be-

gin a literature search is to use electronic databases.

There are many different databases available for

searching and it is important that the appropriate

databases are searched, depending on the objective

and topic of the paper.

Unfortunately, many people

think that only searching MEDLINE is adequate, but

it is not.

This is especially true in when writing on

topics pertaining to chiropractic and allied health

disciplines because many journals from these pro-

fessions are listed, or indexed, in databases other

than MEDLINE. It is usually necessary to search at

least two databases appropriate for the area of study

in order to provide a reasonable breadth and depth

on a topic. Searches through the references of ar-

ticles that are retrieved, authoritative texts, personal

contacts with experts, and reviews of unpublished

primary research may also be warranted and impor-

tant to include.

The breadth and depth of search-

ing is dependent on the topic and objective. Sum-

marily, the author should look in all the locations

that are appropriate for finding the information they

need.

Examples of common databases used in the health

sciences and their areas of focus are listed in table 5.

Some of these databases are free of charge while

others are not. If an author wishes to avoid paying

to perform searches in some of these databases, they

are often available in health science libraries for

students and alumni to use at no cost. Consult a

local chiropractic or other health science library to

find out which databases they carry and also to

inquire about search fees.

Authoritative texts may

also be found at college libraries or even in your

own personal library. At college libraries there are

usually computerized databases that catalog all of

the books in the library’s holdings. These too can be

searched using search terms.

It is crucial to divulge the databases that were

searched in the article.

3,7,13

This means that it is

important for the author to keep track of the

databases searched and the terms used, in order

to report them to the readers. Some authors

also like to keep track of how many ‘hits’ or

article citations that are found with each search. A

sample tracking sheet is provided in table 6. Mini-

mum requirements for narrative reviews are that

authors should state the database searched, a

starting year, and the ending year and month

of the search. The following example is adequate,

“MEDLINE was searched using the terms ‘low back

pain’ and ‘manipulation’ from 1966 through June,

2000.” Stating, “We searched MEDLINE from 1980

Table 4

Standard Subsections of the Narrative Overview Structured Abstract

Subsection Description

Objective • Clearly state the purpose of the paper.

Background • Describe what prompted this review or why it is being written.

• Briefly present a context for the overview.

Methods: • Briefly describe the methods used to conduct the ovewview.

Discussion: • Describe what information the overview presents to the reader.

Conclusion • Summarize what the overview contributes to the literature.

• State what new conclusion is drawn as a result of the synthesis of the

literature reviewed.

Key words • Use terms found in the

Index Medicus

database, which are called medical

subheadings (MeSH).

MeSH can be found at the PubMed home page

(http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=mesh).

• List additional words that may be unique to the topic or to the profession.

Fall 2006

• Number 3 • Volume 5

107

to the present.” is grossly insufficient because

the reader is not told what the word “present”

actually means. Months can slip away between the

day of the search and publication, so it must be

defined.

Search terms & delimiting: Setting the specific param-

eters for the literature search is necessary in order to

make the project feasible since it is not reasonable to

review every single paper that has even the most

minute relation to the topic of study.

17,24

The

boundaries set in this step must be comprehensive

enough to insure that the author may retrieve all

relevant studies, but narrow enough to focus the

effort.

13,17

To do this, one should take the primary concepts or

themes of the topic under study and turn them into

Table 5

Databases to Consider When Performing a Literature Search

Name of Database How to Access Internet Site Pay Site?

PubMed Online software http://www.ncbi.nlm.nih.gov/entrez/

query.fcgi

Emphasis: medicine, nursing, dentistry, veterinary medicine, and preclinical sciences.

Also includes citations for chiropractic, complementary medicine, and allied health.

Index to Chiropractic Online software http://www.chiroindex.org No

Literature Emphasis: chiropractic publications

MANTIS Online search software or

CD-ROM

http://www.healthindex.com Yes

Emphasis: Health promotion, prevention, and conservative alternative therapies,

including: acupuncture, allopathic medicine, chiropractic, herbal medicine,

homeopathy, naturopathy, osteopathic medicine, physical therapy, traditional Chinese

medicine.

ERIC (Educational Online search software http://www.ericsp.org/ No

Resources

Information Center)

Emphasis: Contains more than one million abstracts of education-related documents

and journal articles, including health sciences.

AMED (Allied and

Complementary

Online search software,

CD-ROM, disc, print

http://www.bl.uk/collections/health/

amed.html

Yes

Medicine Database) Emphasis: Unique bibliographic database produced by the Health Care Information

Service of the British Library. Journals related to complementary medicine,

physiotherapy, occupational therapy, rehabilitation, podiatry, palliative care.

CINAHL (Cumulative Online search software http://www.cinahl.com Yes

Index to Nursing and

Allied Health

Literature)

Emphasis: Citations for nursing, allied health, complementary medicine, including

chiropractic.

EMBASE/Excerpta

Medica

Online search software,

CD-ROM, print

http://www.elsevier.com/wps/find/

bibliographicdatabasedescription

.cws_home/523328/

description#description

Yes

Emphasis: Diverse database including: biomedical sciences, human medicine, health

policy and management, occupational and environmental health, alternative and

complementary medicine, nursing.

SPORT Discus Online software vendors http://www.sirc.ca/products/

sportdiscus.cfm

Yes

Emphasis: Sports medicine, physical education, physical fitness, sport law,

administration, training, sport science, kinesiology, coaching, physical therapy,

recreation.

Current Contents Online search software http://scientific.thomson.com/products/

ccc

Yes

Emphasis: Clinical medicine; life sciences; agriculture; biology and environmental

sciences; arts and humanities; physical, chemical and earth sciences; engineering,

computing and technology; social and behavioral sciences.

Cochrane Database of

Systematic Reviews

Online search software http://www.cochrane.org/reviews/en/ Yes

Emphasis: High quality evidence to inform people providing and receiving care, and

those people responsible for research, teaching, funding and administration at all

levels.

DARE (Database of

Abstracts and

Reviews of

Effectiveness)

Online search software http://www.york.ac.uk/inst/crd/

crddatabases.htm

Emphasis: Database of abstracts of systematic reviews assessed for effectiveness.

Volume 5 • Number 3 •

Fall 2006

108

single words, which can be used as search terms. In

addition, insure that the search terms used are rec-

ognized by the vocabulary of the computer data-

base. This is usually easy to do since most comput-

erized indices have a key word search function

within them that will find synonyms in its database

for the word that you enter. Most journals use the

MeSH vocabulary system adopted by the National

Library of Medicine.

These key words will be the

ones that you will also list in the key terms section

of the abstract.

It is not necessary to search databases using solely

key words of the database. Most databases will allow

you to search using additional words that you find

appropriate. This is sometimes helpful in finding

more articles since the databases usually search

through title, abstract and key words of the articles

indexed. If the right terms are not used during the

literature search, then chances are high that some

important studies will be missed. General terms,

such as ‘pain’ may result in so many articles to

search through that the authors miss studies. Like-

wise, highly focused words may narrow the search

down too far, revealing no research. Just as with

reporting the databases used, each search term used

to conduct the search ought to be divulged to the

reader.

How terms are connected together would

also be useful to know.

When writing papers using terms unique to a pro-

fession, it is important to recognize that the term

may have a different meaning in various database.

For example, the term “subluxation” in the medical

literature has a distinctly different meaning than in

the chiropractic literature. The spelling of words

should also be considered, based upon the data-

base being searched. Exemplary is that in the Index

to Chiropractic Literature database, a search using the

spelling “technique” will provide different results

than a search with “technic”. This occurs because

the chiropractic profession historically used the

latter term to describe different chiropractic ad-

justing procedures and the indexing system con-

tinues to use this term. This is also an excellent

example of why it is important to verify the key

words vocabulary of each database. Therefore, to

properly delimit a literature search, authors need

to select key words, including MeSH, keep track

of the terms and search strings used, and report

this information to the reader when writing the

paper.

Selection criteria employed: It is important to briefly

describe what selection criteria were used to include

or exclude a study from the review.

7,10,11,18

This

helps keep the paper focused and helps to insure

that papers are included because of their relevance

to the topic rather than how much the author

agrees or disagrees with the study.

13,18

Exclusion

criteria should be identified that the authors used to

eliminate studies from consideration that were not

pertinent to the focused purpose of the study. Rea-

sons for exclusion may be old data (early research)

or inappropriate topics.

For example, researchers

studying the use of isometric muscle endurance of

the lumbar spine would exclude papers found that

discuss standard orthopedic muscle testing of the

low back because it is not the type of muscle testing

being reviewed.

Inclusion criteria should tell the reader what factors

the authors considered in order to include a paper

in the review. This should include articles pub-

lished in various languages, and other factors perti-

nent to the purpose of the paper. Be careful not to

place too many limiting exclusion criteria or have

inclusion criteria that are too wide; papers outside

the domain of the purpose of the study may be

included inadvertently or inappropriately if this oc-

curs.

Summarily, when writing the methods section, the

author should ask him or herself, “Can the reader

replicate the search that was done based upon what

I have written in the methods section?” If the an-

swer is “Yes” then this section has been adequately

written.

Table 6

Sample Literature Search Tracking Sheet.

Date of Search Database Years Searched Search Terms Strings of Terms # Hits

1/12/99 PubMed 1975–12/1998 Health promotion None used 11330

1/12/99 PubMed 1975–12/1998 Health promotion;

chiropractic

Health promotion and

chiropractic

1/12/99 Cinahl 1966–12/1998 Preventive medicine,

chiropractic

Preventive medicine

and chiropractic

Fall 2006

• Number 3 • Volume 5

109

Discussion

Synthesis: The part of conducting a narrative over-

view that takes the most mental energy is synthe-

sizing all of the information retrieved in the litera-

ture search into comprehensive paragraphs. Since

this is the primary use of reviews of the literature it

is imperative that this section be written as clearly

and as objectively as possible.

It is here that read-

ers should find the information that they want in

one location.

How to structure this section and summarize the

information into a comprehensible synthesis de-

pends on what is being reviewed. If an author is

writing a review pertaining to the current best ap-

proaches in assessing and managing patients with a

particular disorder, it may be a good idea to write

the synthesis in the order that the clinical encounter

normally takes place (eg, history, examination, spe-

cial studies, management). As another example, for

this paper we synthesized the literature in the order

that we felt readers would find it most useful when

they decided to write their own overview. There is

no single way to write this section. Therefore it

requires the author to think clearly about what is

being conveyed according to the objective of the

overview.

Before attempting to write the synthesis authors

should read through each of the papers that will be

included in the overview and take notes on each

one. Most authors prefer to use a word processing

system for taking notes because it is simple to add

new information. Once all of the notes are com-

plete, authors can then easily organize common

themes together. Other authors prefer to take notes

on index cards since they can be shuffled around

and notes can be taken regardless of whether or not

the author is near a computer.

It is recommended that notes include the following:

the purpose of the study reviewed, a synopsis of the

content, the research design or methods used in the

study, a brief review of the findings.

9,17

If an author

plans to objectively evaluate, or critically appraise,

each article, then it is also a good idea to take notes

for this part as well. Regardless of the technique that

you use it is essential to always write the complete

reference down for each set of information that you

extract from a study. It is horrifying to find out late

in the writing process that a reference for a bunch of

information was lost and have to spend hours trying

to find it.

Tables of information may make it easier for some

authors to organize their thoughts when construct-

ing the synthesis.

17,24

Tables are easy to make and

categorize information by topic. For example, the

references for each paper reviewed can be written

down one side of the paper and categories of infor-

mation extracted from each paper can be written

along the top. This is best done using a computer, as

it allows for easy arrangement or additions of infor-

mation. An example of part of a table is presented in

table 7. If papers under review contain a lot of data

that the author feels would be beneficial to include

in the manuscript, then data can be collected using a

table and easily be placed in the final paper.

Table 7

Example of a Part of an Evidence Table

Reference Sample Test Design Results Comments

Biering-Sorensen,

1984

Denmark; 30–60

yo, 449 men,

479 women

Sorensen’s

method

Postal survey 12

months after

exam

Prognostic for

time LBP in

men; women

had insig.

Opposite trend

Some women

omitted from

analysis

Gibbons et al.,

1997

Finland, 35–63

yo, 43 men who

had no LBP

Modified

Sorensen’s

Compared static

back

endurance

time between

group with no

LBP to group

with incidence

of LBP in

preceding 12

mo.

Men who

reported LBP

had slightly

shorter times;

they did not

come close to

stat.

significance.

No association

between static

back

endurance and

future LBP

Difference in

groups may

suggest either

a difference in

condition of

subjects, or a

research

error

Volume 5 • Number 3 •

Fall 2006

110

When writing the synthesis for the manuscript,

the author integrates the findings elicited from the

note taking or table making process. It is important

for authors to keep the reading audience in mind

when writing this section; the audience is less

familiar with the topic under review, necessi-

tating a clear, clinically relevant and easy to read

synthesis.

Major areas of agreement and disagreement in the

literature should be discussed.

The discussion

should tie the study into the current body of lit-

erature, provide its clinical significance, and make

logical interpretations from the literature re-

viewed.

1,18,24

If there is no discussion of the rel-

evance of the overview to other work in the field, it

may signal that the author has not thoroughly in-

vestigated the topic. Since the synthesis is the crux

of the overview design it is important to insure that

a meaningful integration is accomplished.

The author’s interpretation of the literature should

also be provided;

it is for this reason that it is

helpful if the author has some content expertise in

the area under study.

Often times it is appropriate

for the author to offer some critical appraisal of the

papers reviewed in the discussion.

9,11,17

As men-

tioned earlier, this is an area of overviews that some

authorities feel may or may not be necessary in an

overview. One of the reasons for this disagreement

is that the appraisal process can be extremely sub-

jective without the use of explicit critical appraisal

criteria. Since the narrative overview already in-

cludes the biases of the author, there is a limit to

how much more bias may actually enhance the

credibility of the overview.

It is helpful if the author uses check sheets or guides

when critically appraising the articles. These check

sheets are available from a wide variety of sources.

Some of these critical appraisal aids focus on specific

research designs while others are made for assess-

ment of articles in general. In addition to many

worksheets available in the references used for this

paper, a small sample of additional sources for criti-

cal appraisal check sheets are identified in table 8.

When performing critical appraisal, it is best to use a

check sheet that is specific to the research design

being appraised. For example, Appendix A in this

article is a check sheet for a narrative overview of

the literature. A case report or clinical trial should

be reviewed using a completely different check

sheet.

Limitations to the overview: Authors should address

weak points of their own study and mention areas

for improvement.

1,3

No paper is perfect. If limita-

tions are not included in an honest fashion, it warns

Table 8

Examples of Critical Appraisal Check Sheets

Reference Topic Internet Address

Centre for Evidence Based Medicine Articles on therapy, systematic

reviews, diagnosis, prognosis,

harm/etiology, economic analysis

http://www.cebm.net/using_ebm.asp

Critical Appraisal Skills Programme Randomized controlled trials,

economic analyses, qualitative

research

http://www.phru.nhs.uk/casp/

critical_appraisal_tools.htm

Greenhalgh T. How to read a paper:

Papers that summarise other

papers (systematic reviews and

metaanalyses). BMJ 1997;315:672–

675.

Systematic reviews http://www.bmj.com

Oxman AD. Systematic reviews:

Checklists for Review Articles.

BMJ 1994:648–651.

Systematic reviews http://www.bmj.com

Jackson GB. Methods for integrative

reviews. Review of Education

Research 1980;50:438–460.

Integrative reviews of the

literature

Cooper H. Scientific guidelines for

conducting integrative research

reviews. Review of Education

Research 1982;52:291–302.

Integrative reviews of the

literature

Green BN, Johnson CD. Writing

better case reports. J Sports

Chiropr Rehabil 2000;14(3):46–47.

Case reports NA

Fall 2006

• Number 3 • Volume 5

111

the reader that the paper may contain more bias

than is acceptable for an overview. The inability to

discuss the merits and demerits of a paper may

demonstrate that the author has strayed from the

focus of their purpose.

Because authors are deeply involved in the creation

of the literature review and the writing process, it is

not always easy to recognize the paper’s limitations.

One method that is helpful in writing this section is

to write it as the study evolves. Every time the

author recognizes an area that could be improved in

the study, he or she can simply write those thoughts

down at the time. If the weakness can be corrected

during the research process then the writing related

to it can be deleted. For limitations that are inherent

to the design of the paper or ones that are not

correctable, then this text can remain for the final

paper.

Another method that most experienced authors use

is to have trusted colleagues review the paper before

it is submitted to a journal. It is important to select

colleagues who have expertise in the topic under

study as well as those who have writing experience.

These peers will usually be able to see areas of

weakness not immediately apparent to the author

and they can provide suggestions for improvement.

These suggestions can then be incorporated into the

paper and help provide comments for the discus-

sion. Be sure to select assistants who will be honest

in their feedback, will provide the comments in a

timely manner, and who will respect the confiden-

tiality of your work. In receiving feedback from

peers, it is important for the author not to take the

criticism personally, but to see it as a method for

improving the paper.

Conclusion

The conclusion should provide a tie in to the pur-

pose, the major conclusions drawn from the over-

view, and directions for future research.

A clear and concise summary of the major findings

of the overview should be provided.

This is not

merely a rehash of the entire paper, but a statement

about what is now known as a result of the publi-

cation of the overview that was not known or ob-

served before. The conclusion should be drawn from

and supported by the papers reviewed;

1,10,14,18

the

absence of systematic methods should temper the

conclusion.

Specific implications to the practice en-

vironment should also be mentioned.

Authors who

derive conclusions that are irrelevant to the initial

purpose have lost the focus of the review and may

inadvertently infuse bias into the study. If any major

conclusion is not supported by the literature synthe-

sis, then it is a faulty conclusion and your suspicion

about the validity of the paper should be height-

ened. Therefore, conclusions that are made must be

supported by the literature reviewed.

Specific directives for new research initiatives

should be proposed. After reviewing the literature

on a topic the author possesses a vantage point that

may provide valuable guidance for future research

endeavors. Suggestions for new areas of inquiry and

specific study designs are an important outcome of a

thorough review; it is often from these recommen-

dations that researchers begin new studies.

Acknowledgements

It may be appropriate to acknowledge the work of a

colleague who has assisted the author in the prepa-

ration of the manuscript, such as a proofreader or a

person who has provided ideas for the manuscript.

These people can be acknowledged briefly in this

section. People mentioned in this section must give

consent for their name to appear in print, which can

be obtained by asking them to sign a brief statement

that they know that their name will be listed in the

acknowledgements section of the article.

For more

information on the protocol for writing acknowl-

edgements, see the Uniform Requirements.

References

References are an absolute necessity for any re-

search paper, but especially for a review of the

literature. It is extremely important that authors cite

each of the studies reviewed in order to demon-

strate exactly what research was appraised.

All of

the papers included in the review should be refer-

enced. Authors should also cite all supporting re-

search used to write the report. References used to

support the work should come from peer-reviewed

journals, texts, government documents or confer-

ence proceedings.

For the majority of literature

reviews, the use of magazines is not appropriate

because these periodicals are not peer reviewed and

the articles in them are not written with the same

scientific rigor as peer reviewed journal articles.

References should be formatted appropriately. In-

structions for how to write out the references appro-

Volume 5 • Number 3 •

Fall 2006

112

priately for a given journal are usually found in the

journal’s instructions for authors or in the Uniform

Requirements. Proper formatting of references is es-

sential, as it costs time and money on behalf of

journal staff members to send this information back

to authors for correction.

All the information

needed to correctly list a reference can usually be

found with the abstract when conducting a litera-

ture search, or on the pages of the actual journal

article.

Tables

Tables are lists of information that aid in visually

presenting information in an appealing manner

rather than listing information as text in a para-

graph. Such a table in an overview may be the

extraction table used during the synthesis. Tables

should be simple and self-contained, needing no

further explanation. If authors wish to use previ-

ously published tables, the publishing company of

the original material must grant permission and it is

the authors’ responsibility to receive this permis-

sion. Appropriate formatting for tables can be found

in the Uniform Requirements.

Figures

Figures or illustrations are a necessity to make ar-

ticles interesting to read. Since an overview is a

review of text, it is especially useful to use pictures

and tables in order to keep the paper interesting to

read. Most people do not like to read an article that

is nothing but text from the beginning of the title to

the last letter of the references. Pictures can also

help make the paper easier for readers to under-

stand. If authors wish to use previously published

photographs or illustrations, permission must be

granted by the publishing company of the material

and it is the author’s responsibility to receive this

permission.

Complete requirements for preparing illustrations or

photographs for submission are detailed in the Uni-

form Requirements. Captions for each figure used in

the manuscript should be provided. Authors should

not expect that editors will write the figure cap-

tions.

Some journals accept electronic images. Be

sure to scan images at a sufficient resolution to

enable quality printing in the journal. Find out from

the journal which formats are accepted, such as tiff

or jpg files, and whether the journal accepts PC or

Mac formatting.

CONCLUSIONS

Narrative overviews can be a valuable contribution

to the literature if they are prepared properly in an

effort to minimize the author’s biases. Further read-

ing on reviews of the literature is available in a

suggested bibliography located in Appendix B of this

article. Authors wishing to submit narrative over-

views should find this article useful in constructing

such a paper and carrying out the research process.

Given the controversy surrounding the review of

the literature research design, it is our aspiration

that this article will facilitate some scholarly dialog

in the pursuit of creating more valid reviews of the

literature and striking a balance between a the un-

systematic overview and the complex meta-

analysis.

REFERENCES

1. Helewa A, Walker JM. Critical evaluation of research in physical re-

habilitation: toward evidence-based practice. Philadelphia: W.B.

Saunders Co.; 2000. p. 109–124.

2. Day RA. How to write and publish a scientific paper. 5

ed. Phoenix,

AZ: The Oryx Press; 1998. p. 163–167.

3. Crombie IK. The pocket guide to critical appraisal. London: BMJ Pub-

lishing Group; 1999. p. 23–29, 56–62.

4. Gray JAM. Evidence-based health care: how to make health policy and

management decisions. NY: Churchill Livingstone; 1997. p. 72–77.

5. Lang G, Heiss GD. A practical guide to research methods. Lanham, MD:

University Press of America; 1998. p. 17–21.

6. Mendelson T. Keeping up with the medical literature. In: Friedland DJ,

editor. Evidence-based medicine: a framework for clinincal practice.

Stamford, CT: Appleton & Lange; 1998. p. 145–150.

7. Gehlbach SH. Interpreting the medical literature. NY: McGraw-Hill,

Inc.; 1993. p. 243–251.

8. Jadad AR, Cook DJ, Jones A, Klassen TP, Tugwell P, Moher M, et al.

Methodology and reports of systematic reviews and meta-Analyses:

a comparison of Cochrane reviews with articles published in paper-

based journals. JAMA 1998;280:278–80.

9. Portney LG, Watkins MP. Foundations of clinical research: applications

to practice. 2

ed. Upper Saddle River, NJ: Prentice Hall, Inc; 2000.

p. 127–133.

10. Slavin RE. Best evidence synthesis: an intelligent alternative to meta-

analysis. J Clin Epidemiol 1995;48:9–18.

11. Hutchinson BG. Critical Appraisal of Review Articles. Can Fam Physi-

cian 1993;39:1097–102.

12. Sackett DL. Applying overviews and meta-analyses at the bedside. J

Clin Epidemiol 1995;48:61–6.

13. Oxman AD, Guyatt GH. Guidelines for reading literature reviews. CMA

J 1988;138:697–703.

14. Oxman AD, Guyatt GH. Validation of an index of the quality of review

articles. J Clin Epidemiol 1991;44:1271–8.

15. Dixon RA, Munro JF, Silcocks PB. The evidence based medicine work-

book: critical appraisal for clinical problem solving. Woburn, MA:

Butterworth-Heinemann;1998. p. 148–166.

16. Oxman AD, Cook DJ, Guyatt GH. Users’ guides to the medical litera-

ture, VI. How to use an overview. JAMA 1994;272:1367–71.

17. DePoy E, Gitlin LN. Introduction to research: multiple strategies for

health and human services. St. Louis: Mosby-Year Book, Inc; 1993.

p. 61–76.

18. Oxman AD. Systematic reviews: checklists for review articles. BMJ

1994;309:648–51.

Fall 2006

• Number 3 • Volume 5

113

19. Chalmers I, Altman DG. Systematic reviews. London: BMJ Publishing

Group; 1995.

20. Mulrow CD. The medical review article: state of the science. Ann In-

tern Med 1987;106:485–488.

21. Ball C, Sackett D, Phillips B, Haynes B, Straus S. Levels of Evidence and

Grades of Recommendations. Centre for Evidence Based Medicine

(cited 1999 Nov 18]. Available from: URL: http://cebm.jr2.ox.ac.uk/

docs/levels.html

22. Friedland DJ, et al. Evaluating integrative literature. In: Friedland DJ,

editor. Evidence-based medicine: a framework for clinincal practice.

Stamford, CT: Appleton & Lange; 1998. p. 221–46.

23. Sackett DL, Straus SE, Richardson WS, Rosenberg W, Haynes RB. Evi-

dence-based medicine: how to practice and teach EBM. 2

ed. Ed-

inburgh: Chuchill Livingstone; 2000. p. 133–8.

24. Polgar S, Thomas SA. Introduction to research in the health sciences,

ed. Melbourne: Churchill Livingstone; 1995. p. 343–55.

25. Lau J, Ioannidis JPA, Schmid CH. Summing up evidence: one answer

is not always enough. Lancet 1998;351:123–7.

26. Chalmers I. Applying overviews and meta-analyses at the bedside: dis-

cussion. J Clin Epidemiol 1995;48:67–70.

27. Olkin I. Statistical and theoretical considerations in meta-analysis. J

Clin Epidemiol 1995;48:133–46.

28. Lawrence DJ, Mootz RD. Research Agenda Conference 3: editor’s pre-

sentation: streamlining manuscript submission to scientific journals.

J Neuromusculosket Syst 1998;6:161–7.

29. Green BN, Johnson CD. Writing patient case reports for peer reviewed

journals: secrets of the trade. J Sports Chiropr Rehabil 2000;14:51–9.

30. Sackett DL. How to read clinical journals: I. Why to read them and how

to start reading them critically. CMAJ 1981;124:555–8.

31. Lawrence DJ. Structured abstracts and the JMPT. J Manipulative Phys-

iol Ther 1992;15:77–82.

32. Willis JC. Notes for authors. Chiropr Hist 2000;20:5.

Volume 5 • Number 3 •

Fall 2006

114

Appendix A

Narrative Overview Rating Scale

Circle the number that you feel is appropriate for the paper that you are reading:

1 = Absent 2 = Present but not complete 3 = Present and complete

Initial Impression

1 2 3 Does the review appear to be relvant to an issue of interest (18, 30)?

Abstract

1 2 3 Is the specific purpose of the review stated (3, 15)?

1 2 3 Is context for the overview provided?

1 2 3 Is the type of research design stated?

1 2 3 Are the search methods clearly summarized?

1 2 3 Are the important findings clearly discussed?

1 2 3 Are the major conclusions and recommendations clearly outlined?

Introduction

1 2 3 Is the specific purpose of the review clearly stated based upon a brief review of the literature

(1, 3, 18, 24)?

1 2 3 Is the need/importance and context of this study established (2, 11, 24)?

1 2 3 Are novel terms defined (10, 29)?

Methods

1 2 3 Were the electronic databases used to conduct the literature searches identified (MEDLINE,

CINAHL, etc.) (3, 13, 17)?

1 2 3 Were the search years stated?

1 2 3 Were the search terms stated (3)?

1 2 3 Were standard terms used as search terms, including Medical Subject Headings (17)?

1 2 3 Were the guidelines for including and excluding articles in the literature review clearly

identified (10, 18, 22)?

Discussion

1 2 3 Were the results summarized in a comprehensible manner (3, 10)?

1 2 3 Was the critical appraisal of each study the same and reproducible (11, 13, 22)?

1 2 3 Was the quality of the included articles assessed objectively (3, 11, 13)?

1 2 3 Was the variation in the findings of the studies critically analyzed (1, 10, 13, 22)?

1 2 3 Were the meaning of the results addressed (3)?

1 2 3 Do the authors tie in the results of the study with previous research in a meaningful manner

(1, 3, 10)?

1 2 3 Were the weak points and untoward events that occurred during the course of the study

addressed by the authors (1, 3)?

Conclusions

1 2 3 Was a clear summary of pertinent findings provided (10)?

1 2 3 Were the authors’ conclusions supported by the evidence provided (1, 3, 13, 18)?

1 2 3 Were specific directives for new research initiatives proposed?

1 2 3 Specific implications to the practice environment are addressed (3).

Fall 2006

• Number 3 • Volume 5

115

References

1 2 3 Are references relevant, current and appropriate in number (11)?

1 2 3 Are all papers reviewed cited in the references (1)?

Overall Impressions

1 2 3 Do the merits of this review of the literature outweigh the flaws?

1 2 3 Were the authors unbiased in their approach to the review (11, 18)?

1 2 3 Will the results of the paper help me in my philosophical or evidence based approach to

patient care (18, 22)?

Comments & Notes:

Volume 5 • Number 3 •

Fall 2006

116

Appendix B

Additional Suggested Readings on

Literature Reviews

Compiled by Alan Adams, DC, MS, MSEd

Books

Chalmers I, Altman DG. Systematic reviews. Lon-

don: BMJ Publishing Group; 1995.

Cooper H, Hedges LV. The handbook of research

synthesis. New York: Russell Sage Foundation;

1994.

Cooper H. Synthesizing research: a guide for litera-

ture reviews. 3rd ed. Thousand Oaks: Sage Publi-

cations; 1998.

Fink A. Conducting research literature reviews:

from paper to the internet. Thousand Oaks: Sage

Publications; 1998.

Garrard J. Health sciences literature review made

easy: the matrix method. Gaithersburg: Aspen

Publications; 1999.

Goodman C. Literature searching and evidence in-

terpretation for assessing healthcare practices.

Stockholm: SBU: The Swedish Council on Tech-

nology Assessment in Healthcare; 1993.

Light RJ, Pillemer DB. Summing up: the science of

reviewing research. Cambridge: Harvard Univer-

sity Press; 1984.

Mulrow C, Cook D. Systematic reviews: synthesis of

best evidence for healthcare decisions. Philadel-

phia: American College of Physicians; 1998.

NHS Center for Reviews and Dissemination. Under-

taking Systematic Reviews of Research on Effec-

tiveness: CRD Guidelines for Those Carrying Out

or Commissioning Reviews. CRD Report 4, NHS

CRD, University of York, York, UK; 1996.

Oxman AD. The Cochrane Collaboration handbook:

preparing and maintaining systematic reviews.

2nd ed. Oxford: Cochrane Collaboration; 1996.

Journal Articles

Abramson JH. Meta-analysis: a review of pros and

cons. Public Health Rev 1990–91;18(1):1–47.

Clark MJ, Stewart LA. Obtaining data from ran-

domised controlled trials: how much do we need

for reliable and informative meta-analyses? BMJ

1994;309:1007–1010.

Crombie IK, McQuay H. The systematic review: a

good guide rather than guarantee. Pain 1998; 76:

1–2.

Dickersin K, Berlin JA. Meta-analysis: state of the

science. Epidemiol Rev 1992;14:154–76.

Dickersin K, Scherer R, LeFebvre C. Identifying rel-

evant studies for systematic reviews. BMJ 1994;

309:1286–91.

de Vet HCW. Systematic reviews on the basis of

methodological criteria. Physiotherapy 1997;

83(6):284–289.

Egger M, Smith GD, Phillips AN. Meta-analysis:

principles and procedures. BMJ 1997;315:1533–

1537.

Eysenck HJ. Meta-analysis and its problems. BMJ

1994;309:789–792.

Moher D, Cook DJ, Jadad AR, Tugwell P, Moher M,

Jones A, et al. Assessing the quality of reports of

randomised trials: implications for the conduct of

meta-analyses. Health Technol Assess 1999;3(12):

i-iv, 1–98.

Mosteller F, Colditz GA. Understanding research

synthesis (meta-analysis). Annu Rev Public

Health 1996;17:1–23.

Mulrow CD. Rationale for systematic reviews. BMJ

1994;309: 597–599. Rationale for systematic re-

views. Peto R. Why do we need systematic over-

views of randomized trials? Stat Med 1987;6(3):

233–240.

Thacker SB. Meta-analysis: a quantitative approach

to research integration. JAMA 1988;259: 1685–

1689.

Thompson SG. Why sources of heterogeneity in

meta-analysis should be investigated. BMJ 1994;

309:1351–1355.

Zwarenstein M, Volmink J, Irwig L, Chalmers I.

Systematic review: ‘state of the science’ health-

care decision-making. S Afr Med J 1995;85(12):

1266–1267.

Fall 2006

• Number 3 • Volume 5

117