f re

or
)f
om
.c
ow
kn
ed
m
The IMRaD format

w.
w
(w
ow
kn
ed
M
by
Philip Abraham
ed
st
ho

P D Hinduja National Hospital and

te
si

K E M Hospital, Mumbai
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Why IMRaD

.c
ow
kn
ed
m
w.
“The man of science appears to be the only person

w
(w
who has something to say just now, and the only

ow
kn
man who does not know how to say it.”

ed
M
by – Sir James Barrie
ed
st
ho
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
WWWHWAW

.c
ow
kn
ed
m
“I keep six honest serving-men

w.
w
(w
(They taught me all I knew)

ow
Their names are What and Why and When

kn
ed
And How and Where and Who”
M
by
ed
st
ho
te

-- Rudyard Kipling (1865-1936). “The Elephant’s Child”

si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
What is IMRAD?

.c
ow
kn
ed
m
w.
I Introduction

w
(w
w
M Methods

o
kn
ed
R Results

M
a and by
ed
st
ho

D Discussion
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
History

.c
ow
kn
ed
m
1665 Origin of scientific papers

w.
w
(w
w
1600s and 1700s Letters and experimental (descriptive)

o
kn
formats coexisted

ed
M
1800s (second half) Increasing Methods description (“theory
by
– experiment – discussion”)
ed
st
ho

1900s (early) Organized as in book chapters (heading

te

according to subject)
si
a

1900s (second half) Adoption of IMRaD format

m
fr o
d
de
oa
nl
w
 f re
or
)f
om
History in 1900s

.c
ow
kn
ed
m
w.
• Up to 1945 Titles as in book chapters

w
(w
w
• 1950 to 1960 IMRaD structure partially adopted

o
kn
ed
• After 1965 IMRaD began to predominate

M
• 1979 by
IMRaD introduced as standard by
ed
st

American National Standards Institute

ho
te
si

• 1980s Absolute leadership of IMRaD

a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Background

.c
ow
kn
ed
m
w.
• Considered ideal outline in early 1900s

w
(w
• Physics adopted IMRaD in 1950s

o w
kn
• After World War II, international conferences on scientific

ed
publishing recommended IMRaD

M
• by
Late 1970s, International Committee of Medical Journal Editors
ed
st

(“Vancouver Group”) first published guidelines

ho
te

• Wide use of IMRaD may be credited to editors, to benefit

si
a

readers and facilitate peer review

m
fr o
d
de
oa
nl
w
w
nl
oa
de
d
fr o
m
a
si
te
ho
st
ed
by
M
ed
kn
o w
(w
w
w.
m
ed
kn
ow
IMRaD adoption by major journals

.c
om
)f
or
f re
 f re
or
)f
om
Hill’s
Bradford Hill’ s questions

.c
ow
kn
ed
m
w.
w
Introduction Why did you start?

(w
w
Methods What did you do?

o
kn
ed
Results What did you find?

M
and by
ed
st

Discussion What does it all mean?

ho
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Starting a conversation

.c
ow
kn
ed
m
w.
A: What’s news, babe?

w
(w
B: You know that guy Rakesh … done a lot of work on hepatitis E

ow
kn
… I think he’s asked good questions … but, you know what …

ed
M
you and I can find holes in his arguments and come up with a
by
ed

shocker …
st
ho

A: Wow! … tell me more … keep singin’, babe

te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Introduction

.c
ow
kn
ed
m
w.
• Brief and arresting

w
(w
• Define nature and scope of problem, but

w
o
kn
• Do not hide inconvenient facts

ed
M
by
ed
st
ho
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Introduction

.c
ow
kn
ed
m
w.
• Adequate information to allow reader to understand and

w
(w
evaluate present study without referring to previous publications

ow
kn
ed
• Define lacunae and shortcomings in current state of knowledge

M
• by
Key references to support background information provided
ed
st

• Refer to your previous preliminary work and closely related

ho
te

papers appearing elsewhere

si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Introduction: “funneling”
“funneling” down

.c
ow
kn
ed
• Provide rationale for current study

m
w.
w
– What gap in knowledge did you try to fill?

(w
w
– What controversy did you try to resolve?

o
kn
ed
• State aim of study

M
•
by
May briefly state study group, design and methods used,
ed
st
ho

especially why these are better than in previous studies

te
si

• May state principal result/conclusion

a
m

(but this may take away “surprise” element …

fr o
d

oh, well, it’s already out in the Abstract)

de
oa
nl
w
 f re
or
)f
om
Methods

.c
ow
kn
ed
m
w.
The three questions

w
(w
• What has been done?

ow
kn
• What did you look for?

ed
M
• How was it done? by
ed
st

Should be reproducible by another group

ho
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Methods: details

.c
ow
kn
ed
m
• Study design (drug trial / intervention; prospective / retrospective;

w.
w
randomized, blinded; sensitivity of method; questionnaire; case report;

(w
w
guidelines; meta-analysis)

o
kn
• Setting

ed
• Who is the study about?

M
by
– Participants and control subjects (in animal studies, specify genus,
ed
species)
st

• What did you do?

ho

– Intervention
te
si

– Follow up
a
m

• What did you look for?

fr o

– Outcome measure
d
de
oa
nl
w
 f re
or
)f
om
Methods: details

.c
ow
kn
ed
m
• Inclusion criteria

w.
w
(w
• Exclusion criteria

w
• Sample size calculation

o
kn
ed
• Circumstances under which intervention done

M
– Lab settings by
ed

– In-patient or real life

st
ho

• Consent
te
si

• Ethics clearance
a
m
fr o

(Sections and subsections help)

d
de
oa
nl
w
 f re
or
)f
om
Methods: interventions and tests

.c
ow
kn
ed
m
w.
• If standard, give reference

w
(w
• If new or modified, provide details (sufficient for reproduction by

ow
kn
other workers)

ed
M
• Timing and duration of intervention
by
ed
• Equipment / kits / manufacturer
st
ho
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Methods: outcome measurement

.c
ow
kn
ed
m
w.
• Define outcome

w
(w
• Parameters to assess outcome

ow
kn
• Endpoint, cut-off values

ed
M
• Adverse events, if any
by
ed
st
ho
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Follow up

.c
ow
kn
ed
m
• Frequency, method, duration (including minimum acceptable

w.
w
(w
duration)

ow
• Criteria for termination or drop-out

kn
ed
M
– Per-protocol vs. intention-to-treat
by
ed
st
ho
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Statistical analysis

.c
ow
kn
ed
m
w.
• Methods used for different parameters

w
(w
• Software

ow
kn
ed
M
by
ed
st
ho
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Methods: general

.c
ow
kn
ed
m
w.
• Sub-headings should be consistent with those of Results

w
(w
w
• Try to avoid more than 3 levels of heading

o
kn
ed
M
by
ed
st
ho
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Results: general

.c
ow
kn
ed
m
• What did you find?

w.
w
(w
• Should answer all points raised in Methods

ow
kn
• No new parameters

ed
M
• No mismatch in numbers between text and tables / figures
by
ed
st
ho
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Results: participants

.c
ow
kn
ed
m
w.
• How many screened?

w
(w
• How many eligible?

ow
kn
• How many recruited / excluded?

ed
M
• How many completed study?
by
ed

• Reasons for lack of completeness

st
ho

• Compliance with therapy / protocol

te
si
a

All subjects should be accounted for

m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Results: data presentation

.c
ow
kn
ed
m
• Cause of incomplete data, if any (sample lost, incomplete study)

w.
w
(w
• No repetition between text and tables

ow
• No interpretation

kn
ed
• No adjectives (most, some, often..)

M
• Use % only if n>100 by
ed
st

• Restrict decimal points to 1 or 2

ho
te

• Provide value of p (“highly significant”, “very highly significant”

si
a

meaningless)
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Discussion: outline

.c
ow
kn
ed
• Recapitulation of major findings

m
w.
• Discussion of major findings in light of available data

w
(w
• Discussion of important minor findings

ow
kn
• Alternative explanations

ed
M
• Strengths and limitations of study
by
ed

• Implications of findings
st
ho

• Unanswered questions and future research

te
si
a

• Summary / conclusion
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Common mistakes: Introduction

.c
ow
kn
ed
m
w.
• History starting from Adam

w
(w
• Details of previous studies

w
o
kn
• Aggrandizement

ed
M
• Abbreviations without full form
by
ed

• Details of Results and Conclusions

st
ho

• Intermix with Discussion

te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Good Introduction

.c
ow
kn
ed
m
w.
“We wish to suggest a structure for the salt of deoxyribose nucleic acid

w
(w
w
(D.N.A.). This structure has novel features which are of considerable

o
kn
ed
M
biological importance.”
by
ed
st
ho

-- Watson JD, Crick FHC. A structure for deoxyribose nucleic acid. Nature
te
si

1953;171:737-8
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Common mistakes: Methods and Results

.c
ow
kn
ed
m
w.
• Mixed up

w
(w
• Errors in data (e.g., mean age 25, range 17-22)

ow
kn
• Mismatch of data in Methods / Results / Tables / Figures

ed
M
• Misinterpretation of data by
ed
st
ho
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Common mistakes: Discussion

.c
ow
kn
ed
m
w.
• First study in the world / India / Maharashtra…(megalomania)

w
(w
• Repeating results

wo
kn
• Emphasizing strengths of study over its weaknesses

ed
M
• by
Inflating importance of findings
ed
st

• Going beyond evidence and drawing unjustified conclusions

ho
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Benefits of IMRaD

.c
ow
kn
ed
m
w.
• Development and changes in internal organization of scientific

w
(w
article is answer to constant growth of information

w
o
kn
• IMRaD structure facilitates modular reading

ed
M
• Readers usually do not read in linear way but browse in each
by
ed

section of article, looking for specific information, which is

st
ho

normally found in pre-established areas of the paper

te
si
a

-- Meadows. J Inf Sci 1985;11:27-30

m
fr o
d
de
oa
nl
w
 f re
or
)f
om
You thought IMRaD was gospel?

.c
ow
kn
ed
m
w.
• Nature Medicine prints Methods last and in smaller type

w
(w
• Science buries explanatory footnotes within reference list

w
o
kn
• Lancet editor referred to “…shaky pillars of IMRaD”

ed
M
• by
IMRaD suggests perfectly planned and beautifully executed
ed
st

projects free from accidents and human error

ho
te

• IMRaD does not tell writer how much to put in or leave out or
si
a

what level of reader to aim at

m
fr o
d
de
oa
nl
w
 f re
or
)f
om
Sections not covered by IMRaD

.c
ow
(but covered by Kipling
Kipling))

kn
ed
m
w.
Title How long; how many parts; declamatory (or not)?

w
(w
Authorship Who is best defined in advance; what does

ow
kn
"authorship" mean; how many?

ed
M
Summary What structure; where to place it; how long?
by
ed

Conclusion Who needs one?

st
ho

Acknowledgments Who should be thanked; who paid; who has conflicts?

te
si
a

References How many; what are they for; how to set them out?
m
fr o
d
de
oa
nl
w
 f re
or
)f
International Committee of Medical Journal Editors,

om
.c
ow
not

kn
ed
Immaculately Conceived Moses, Jesus, Et al

m
w.
w
(w
ow
kn
ed
M
by
IM R aD
ed
st
ho
te
si
a
m
fr o
d
de
oa
nl
w
 f re
or
)f
om
An alternative to IMRaD

.c
ow
kn
ed
m
• Brief description of context

w.
w
(w
• Outline of problem

w
• Key measures for improvement

o
kn
ed
• Process of gathering information

M
• Analysis and interpretationby
ed

• Strategy for change

st
ho

• Effects of change
te
si

• Next steps
a
m

-- Br Med J 2000;321:1428
fr o

(recommended for Quality Improvement Reports)

d
de
oa
nl
w
w
nl
oa
de
d
fr o
m
a
si
te
ho
st
ed
by
Abide by Instructions, but

M
ed
kn
o w
(w
w
w.
m
ed
kn
ow
.c
om
)f
or
f re
 f re
or
)f
Abide by Instructions, but a little liberty is sometimes in order

om
.c
ow
kn
ed
m
w.
w
(w
ow
kn
ed
M
by
ed
st
ho
te
si
a
m
fr o

Scientific communication need not be oh so boring!

d
de
oa
nl
w