Answer the questions

NotAllF

Article4Q4.pdf

Home >Social Science homework help >Answer the questions

Singapore Med J 2010; 51(5) : 362

Effective Medical Writing Pointers to getting your article published

Ng K H, Peh W C G

CME Article

Writing a systematic review

ABSTRACT

Evidence-based medicine (EBM) aims to combine

the best available scientific evidence with clinical

experience and individual judgment of patient

needs. In the hierarchy of scientific evidence,

systematic reviews (along with meta-analyses)

occupy the highest levels in terms of the quality

of evidence. A systematic review is the process

of searching, selecting, appraising, synthesising

and reporting clinical evidence on a particular

question or topic. It is currently considered

the best, least biased and most rational way to

organise, gather, evaluate and integrate scientific

evidence from the rapidly-changing medical and

healthcare literature. Systematic reviews could

be used to present current concepts or serve

as review articles and replace the traditional

expert opinion or narrative review. This article

explains the structure and content of a systematic

review.

Keywords: evidence-based medicine, medical

writing, systematic review

Singapore Med J 2010; 51(5): 362-366

INTRODUCTION Evidence-based medicine (EBM) integrates the best

scientific evidence with clinical expertise and individual

judgment of patient needs. While the evidence from one

scientific trial may refer to a particular method or practice

as the best in current clinical practice, the evidence may

change over time as new treatments or methods are

developed. EBM is based on clinical evidence derived

from randomised controlled trials, systematic reviews,

meta-analyses and clinical guidelines. The hierarchy of the

strength of evidence for decision-making and treatment

options flows from the top ranking of randomised controlled

trials, to systematic reviews of these randomised trials, all

the way down to unsystematic clinical observations.(1,2)

Therefore, systematic reviews and meta-analyses are

Biomedical Imaging and Interventional Journal, c/o Department of Biomedical Imaging, University of Malaya, Kuala Lumpur 50603, Malaysia

Ng KH, PhD, MIPEM, DABMP, Editor

Singapore Medical Journal, 2 College Road, Singapore 169850

Peh WCG, MD, FRCP, FRCR, Advisor

Correspondence to: Prof Ng Kwan Hoong Tel: (60) 3 7949 2069 Fax: (60) 3 7949 4603 Email: [email protected]. my

regarded as popular evidence-based tools for analysis of

clinical studies, and they are frequently used to answer

complex research questions across many different clinical

fields.(3) Meta-analysis is for the specific statistical technique

of combining the data from individual studies.

A systematic review is a rigorous way of summarising

the available scientific evidence that is derived from several

clinical trials on a particular treatment or method. It uses a

methodology of clearly-designed questions and methods to

identify and critically evaluate relevant research, followed

by the collection and analysis of data from the studies that

are included in the review. Since the results of a single

study apply only to a certain type of patient or a particular

clinical setting, a systematic review of many studies on the

same subject can provide information that is relevant to a

broad range of patients across different clinical settings.

Moreover, individual studies consist of different sample

sizes, thus biased opinion is inevitably introduced into the

studies. A systematic review limits bias while improving

the reliability and accuracy of recommendations because

it combines information from individual studies and has an

overall sample size that is greater than that of any one study.

This leads to an increase in the quality of the review. In

short, a systematic review is currently considered to be the

best, least biased and most rational way to organise, gather,

evaluate and integrate scientific evidence from the rapidly-

changing medical and healthcare literature.

STRUCTURE AND CONTENT OF A SYSTEMATIC REVIEW A systematic review follows the same structure as an

original research article.

Title: The title of the systematic review should accurately

reflect the topic under review. Some examples of titles are:

• Is there a role for endorectal balloons in prostate

radiotherapy? A systematic review

• Diagnostic value of non-invasive imaging techniques

in the detection of carotid artery stenosis: A systematic

review

Singapore Med J 2010; 51(5) : 363

• A systematic review of pharmacological

interventions for acute ischaemic stroke

Abstract: A structured abstract that summarises the

background, methods, results and conclusion should

follow.

Introduction: The introduction should summarise the

topic and explain the necessity for conducting such a

systematic review. For example, were there gaps in the

knowledge of the topic, or was there disagreement in the

literature? The introduction should be brief and the last

sentence should clearly state the aims of the review.

Box 1. Structure of a systematic review:

• Title

• Structured abstract

• Introduction

• Methods

• Results

• Discussion

• References (extensive number)

Methods: Explain the review methodology clearly and

logically.

• Inclusion and exclusion criteria: Describe the

studies to be included, e.g. a prospective study

in a particular population testing a particular

treatment. In this section, use subheadings, e.g.

who the patients/subjects/animals were; what the

interventions/characteristics sought were; and what

the outcome measures and study characteristics

were. Mention certain key features of the study such

as the sample size, the main results, whether there

was any language restriction or years within which

the studies had to be done. What were the studies

excluded from the review?

• Identification of studies: Provide details of the

electronic database search carried out, including the

databases used (e.g. MEDLINE, Scopus, ISI Web

of Knowledge, Google Scholar), and the period

of search. Provide a list of the search terms in an

appendix. Keywords used in the searching of studies

should be listed in this section. State whether any

of the journals were identified using hand-searching

(e.g. reference list of articles obtained) and whether

review articles and other bibliographies were

referenced. In most reviews, only studies published

in the English language are included; however, some

researchers may include non-English studies and

this issue should be addressed.

• Study selection: Describe the selection method and

screening for eligibility for all the studies that were

identified, removing duplicates and screening for

relevance of the title, followed by abstract and full

text article; and then screening of the full papers

that passed all of these eligibility checks. A flow

chart is recommended to demonstrate the search

strategy utilised (Fig. 1).

• Data extraction: Describe what data were extracted

from the studies, such as the description of patients/

subjects/animals included, how many had what

outcome, what the summary statistics were as

given in the paper, and whether the data was read

off graphs if these were not provided numerically.

Data should be independently extracted from the

studies by at least two researchers to avoid biased

opinions.

• Quality assessment: Quality assessment criteria

have been described for most literature types and

many of these are available on the EQUATOR

Network (www.equator-network.org).

• Data analysis: Describe briefly the statistical

methods of handling the data. In a review on

treatment, it is typical to calculate the odds ratios

for each outcome with 95% confidence intervals and

p-value for the magnitude of effect, and to perform

tests of heterogeneity to ascertain if the studies were

all coming broadly to the same conclusion or if there

were significant variations between the studies. The

analysis may also be influenced by the amount of

data available. Heterogeneity needs to be assessed

for each outcome. The studies or outcomes can be

sorted into different subtypes to determine whether

this reduces any heterogeneity, for example, in a

systematic review of a treatment, the studies may

be sorted by dose, by time of administration after

the onset of disease or by time of assessment of

outcome.

Fig. 1 Flow chart for identifying eligible articles.

Medical database

121 Articles (Citations retrieved from the database)

51 Excluded (Reading of abstract)

70 Articles 42 Excluded (Do not meet the criteria)

28 Used in the study

Singapore Med J 2010; 51(5) : 364

Results: As with the methods section, the results

section needs to be described logically.

• Search results: Mention the number of studies the

literature search had identified from each database

as well as by other searching methods, how many

were duplicates, how many were excluded on the

basis of screening of an abstract, how many full

text articles were assessed and of these, how many

were excluded until the final number of papers

to be included was determined. Also mention the

number of papers not included in the review due to

publication in the non-English literature.

• Study range and characteristics: Describe the

demographic and descriptive statistics of the

patients/subjects/animals included in the selected

studies. Indicate the type of patients/subjects/

animals that were included, such as an age range

or disease severity range. Indicate other key study

methodological features such as any variation in

outcome measure. Be careful to ensure that the

same patients/subjects/animals have not been

included in multiple publications as this would

artificially inflate the sample size. State the total

number of actual individual studies within the total

number of papers identified. Many of the study

characteristics can be summarised in graphs or

tables.

• Study quality and potential sources: Report the

median quality score derived from the appropriate

quality assessment method and indicate which

studies had particularly poor quality score points.

The study quality score can also be used to see

whether any apparent heterogeneity between

studies for outcome results can be explained by the

inclusion or exclusion of poor quality studies.

• Effect of intervention on outcome: Describe the

actual results of analysis to estimate the overall

effect across the different studies. For example,

describe the odds ratios for each outcome in a

treatment review or the mean differences in an

observational study. Explore any potential reasons

for heterogeneity and describe what that showed.

Discussion: This section should begin with a simple

statement summarising the major findings from the

review. For example, is the weight of evidence in favour

of the treatment having a beneficial effect? Do the studies

suggest that some particular feature is associated with

future risk of a disease? Is one diagnostic test better than

another? Can a less invasive interventional procedure be

reliably used as an alternative to the conventional invasive

procedure? If there is not enough evidence to arrive at a

definite conclusion, then the statement should clarify

that.

The subsequent paragraphs should describe the

limitations of the studies included and the reliability of the

results. How robust are the results? Were there any biases?

Next, describe the strengths and weaknesses of the review

methods. For example, were non-English publications

omitted? Does the review only include a handful of papers

with a small sample size? Put the results in the context of

other knowledge on the topic, for example, by comparing

this review with previously-published systematic reviews

or current opinions and guidelines.

Finally, provide conclusions and any implications

for current practice, and particularly for future research.

Has the review highlighted gaps in knowledge that future

studies should address? Is there enough new information

to modify existing clinical practice?

Like any other research paper, a systematic review

should acknowledge the funding agencies and grants

received, as well as any other persons who helped during

the review. References for the selected studies should

be listed. The use of effective tables and figures would

enhance the readability of the review.

Box 2. Common problems with systematic reviews:

• Important studies are missing due to an inadequate

literature search.

• Too many biases introduced.

• No clear hypothesis or research question.

• Review methodology is not clearly stated.

• No clear conclusion or statement to summarise the

findings of the review.

SUMMARY A systematic review is a process of searching, selecting,

appraising, synthesising and reporting clinical evidence on

a particular question or topic. A quality systematic review

requires substantial preparation and planning before it

is written. After thorough development of the research

question and protocol, a considerable amount of effort is

required to search the literature, appraise the quality of the

selected studies, and finally reach thoughtful, appropriate

conclusions. A systematic review should be written just

Singapore Med J 2010; 51(5) : 365

like any other research paper, with a logical flow of the

methods, results, discussion and conclusion.

Box 3. Key points of a systematic review:

• A systematic review searches, selects, appraises,

synthesises and reports on current clinical evidence

on a particular question or topic.

• A systematic review uses strategies that limit bias and

random error.

• A systematic review helps medical practitioners

keep up-to-date with the rapidly changing medical

literature.

• A systematic review ranks among the highest in the

hierarchy of quality clinical evidence for evidence-

based medicine.

REFERENCES 1. Green S. Systematic reviews and meta-analysis. Singapore Med J

2005; 46:270-3. 2. Guyatt G, Drummond R. Part 1. The basics: Using the medical

literature. 1A. Introduction: The philosophy of evidence based medicine. In: Guyatt G, Rennie D, eds. Users’ Guides to the Medical Literature: A Manual for Evidence-Based Clinical Practice. Chicago. AMA Press, 2002: 3-12.

3. Petrie A, Bulman JS, Osborn JF. Further statistics in dentistry Part 8: Systematic reviews and meta-analyses. Br Dent J 2003; 194:73-8.

USEFUL RESOURCES • EQUATOR Network. Available at: www.equator-network.org. • The Cochrane Collaboration. The reliable source of evidence in

healthcare. Available at: www.cochrane.org. • CEBM Centre for Evidence-based Medicine. Available at: www.

cbem.net • The Campbell Collaboration. Available at: www.

campbellcollaboration.org.

Singapore Med J 2010; 51(5) : 366

SINGAPORE MEDICAL COUNCIL CATEGORY 3B CME PROGRAMME Multiple Choice Questions (Code SMJ 201005A)

Doctor’s particulars: Name in full: __________________________________________________________________________________

MCR number: _____________________________________ Specialty: ___________________________________

Email address: _________________________________________________________________________________

True False

☐ ☐ ☐ ☐

☐ ☐

SUBMISSION INSTRUCTIONS: (1) Log on at the SMJ website: http://www.sma.org.sg/cme/smj and select the appropriate set of questions. (2) Select your answers and provide your name, email address and MCR number. Click on “Submit answers” to submit.

RESULTS: (1) Answers will be published in the SMJ July 2010 issue. (2) The MCR numbers of successful candidates will be posted online at www.sma.org.sg/cme/smj by 2 August 2010. (3) All online submissions will receive an automatic email acknowledgment. (4) Passing mark is 60%. No mark will be deducted for incorrect answers.(5) The SMJ editorial office will submit the list of successful candidates to the Singapore Medical Council.

Deadline for submission: (May 2010 SMJ 3B CME programme): 12 noon, 26 July 2010.

Question 1. A systematic review is aimed at:

(a) Studying the benefits of a treatment method within a small sample size.

(b) A comprehensive narrative analysis of recent developments in a specific topic.

(d) Appraising, synthesising and reporting based on current clinical evidence.

Question 2. The structure of a systematic review includes:

(a) A structured abstract

(b) A list of all selected studies in the methods section.

(d) A results section describing the analysis of the studies.

Question 3.The methods section of a systematic review includes:

(a) Identification of studies.

(b) Study selection.

(d) Quality assessment.

Question 4. When describing the results of the literature review:

(a) State the number of studies identified and from which database.

(b) Describe the study range and characteristics.

(d) Describe the limitations of the studies included and of the systematic review.

Question 5. A quality systematic review should:

(a) Reach appropriate conclusions.

(b) Improve the reliability of recommendations.

(d) Identify gaps in the knowledge.