12 Tips for Sci Paper

8/8/2019 12 Tips for Sci Paper

http://slidepdf.com/reader/full/12-tips-for-sci-paper 1/3

International Journal of Angiology 11:53-55 (2002)DO1:10.1007/s00547-001-0054-8

T w e l v e T i p s o n W r i t i n g a G o o d S c i e n t i f i c P a p e r

J o h n A . E l e f t e r i a d e s , M.D., F.I.CoA.

Department of Cardiothoracic Surgery, Yale University School o f Medicine, New Haven, Connecticut

It is an indication of Dr. Chang's wisdom and experi-ence in academic matters that he should commission an

article on the technique o f writing a good scientific pa-

per. There are many pitfalls to which the new or occa-

sional writer is prone. Also, the manner of presentation

can make a great deal of difference as to how favorably

research information is received by editors and can im-

pact the usefulness of the paper to readers.

The compilation which follows incorporates sugges-

tions which the author hopes will be of use in writing

scientific papers. Much of this was taught to the author

by his mentors, and much was learned (the hard way) by

experience. This list is not meant to be exhaustive, but

rather to cover certain important general principles re-

garding how to present one's research data for scientificpublication.

Good Underlying Research Question

1. A good research paper needs to start with a good

research questio n--a question to which the reader wants

to know the answer. If the underlying question is not

pertinent or interesting or of clinical importance, no one

will want to know the answer, and no one will be in-

terested in accepting or reading the resulting paper. At

what size does an aneurysmal aorta rupture? How often

are infectious pathogens found in atheroma removed atsurgery? Does revascularization improve ventricular

function in patients with ischemic cardiomyopathy?

What is the influence of antecendent pulmonary hyper-

tension on mortality in heart transplantation? Can ar-

teriosclerosis occur in the absence of interferon-y? These

are all examples of research questions in angiology that

produced published papers which proved of interest

to readers. Good basic questions that stimulate

research--both in the laboratory and the clinic--arise

from the investigator's experience and creativity. With-

Correspondence to: John A. Elefteriades, M.D., Professor and Chief,Department o f Cardiothoracic Surgery, Yale University School ofMedicine, 333 Cedar Street, P.O. Box 208039, New Haven, CT 06520-8039

out such worthy underlying questions, the resultingmanuscript will not be of interest to editors, nor will the

paper capture readers' attention. Most often, for clinical

papers, a well-chosen question will impact substantially

on clinical management of real-world patients.

Brevity

2. The paper must be concise. This is often a hard

concept for scientific writers just s tarting out. The new

scientific writer is often not accustomed to the brevity of

scientific communicat ion , which is very different from

other forums. Excess length is not as much an issue for a

thesis or for general prose or fiction. But, the scientificreader is hard-pressed for time. For scientific journals,

space is at a premium. You must state your concept,

methods, results, conclusions, and discussion succinctly,

with as few words, Tables, Figures, and References as

possible. With experience, you will come to realize just

how much can be cut without adversely affecting con-

veyance of your message. I learned this concept from

Dr. Arthur Baue, my former Chairman, who cut a

chapter I had written on diaphragm pacing dramatically

without curtailing its message whatsoever. Dr. Baue,

editor of Glenn's Textbook of Thoracic and Cardio-

vascular Surgery, taught me just how much of what we

put on paper is repetitious or unnecessary to convey our

meaning. I learned this concept also from Dr. William

Roberts, editor of the American Journal of Cardiology.

Dr. Roberts has a reputation for cutting and cutting

without detracting in any way from the substance of a

report. This is a very important talent for the novice

scientific writer to learn.

Abstract Conveys All Critical Information

3. The abstract must convey all the cardinal findings

and messages. Please remember that even the an avid

reader of your work will read your abstract and at

most look at some of your figures, skim y~ur methodsand results, and glance at your discussion and con-



54 J A. Elefteriades: Writi ng a Scientific Pap er

40

gc-

O 30

"6

ee

-1 0

5 6 7 8

Aortic Size (cm)

SIIh~ IFN,-~

" ~ ~

f t ,~s

~-~'1 ar~l I '~e,~ ir,~ rr~rnrmfI (o, ~ aa~/g talts &5 ~ ~W ~ taa ~a ~o n ~

,no .~y ~ LA - ~f~ ~,i~o~4 LV - ~st een~,/c~ IIA - , ight

~t T/ u m ; RV = g h o s t v t ' ~q d ~ D = ~ g - ~ipltnL)

v 3.5 to 3.9 em ~4.0 to 4.9 cm w $.0 Io 5.9 c :n m2 6.0em

II1%

16~ | £ ,6%

,4̀%

4,% -

2,%

C~¢ ¢ o m e

Fig. 1. Examples o f cardinal visual images defining specific published pape rs.

clusions. Very few people--your editor, your reviewers,

and readers working in the exact same area as

you--will actually read the full content of your paper.

Be sure that everything you put in your abstract is

important and that nothing important is left out of

your abstract.

A void Excessive Li terature Review

4. Do not overdo y our review of the literature. Begin-

ning medical writers have a tendency to review past

studies on similar topics exhaustively. This is not nec-

essary and can be excessively cumbersome for editors

and readers. You must recognize prior work, but at the

same time be aware that the purpose of your paper is to

say what you studied and what were your findings.

Whole series of papers can often be summarized by one

or a few sentences describing the general thrust of prior

research. In your discussion, you will be able to address

significant similarities and contrasts vis-/t-vis prior workin a succinct fashion.

A void Excessive Use o f Figures

5. Remember that the Figures appropriate for a publi-

cation are very different from those for a lecture pre-

sentation. For a presentation, you want to illustrate

every point for ease of assimilation. Yo u will want ma ny

slides. By contrast, for a publication, yo u need to min-

imize the illustrations. The reader can assimilate simple

numerical data from the body of the text. Only complex

compilations of data require a Figure or Table. Chose

your figures for publication critically and sparingly.

One Cardinal Visual Image

6. You should have one single, cardinal visual image

which the reader will recall long-term that conveys the

crux of your paper. This is what I call the Wacker's rule

(after Frans Wackers, our distinguished chief of Nuclear

Cardiology at Yale). We all know that visual material is

readily assimilated and visual images easily recalled.Your cardinal figure may convey a change in a certain



J.A. Elefteriades: Writing a Scientific Paper 55

paramet er before or after tr eatment. The figure may show

a germinal histologic finding. The figure may illustrate

the steps of a new technique. Nonetheless, you should

have such a central figure, and it must be an exceptional

one. You should ask yourself, if a colleague were to cite

my work in a lecture, wha t one slide would I want him toshow. This will be your cardinal figure. (Some illustra-

tions of "cardinal figures" from publications by our

group are included in the accompanying Figure.)

series" or "shows the lowest mortality," or other such

superlative self-assessments. Very often, som eone from

somewhere will disprove your claim-perhaps from an

article published in an obscure journal from another

country. Your work and your data will speak for

themselves. I learned this from the very wise Dr. WilliamW. L. Glenn, the great cardiac surgical pioneer from

Yale University.

C a r e f u l F i n a l R e - R e a d i n g

7. Read and re-read until you know that everything in

your paper is accurate. Make sure not only that all your

data is well-substantiated and scientifically accurate, but

also that all the numbers "add up." Usually it is best to

have at least one of your co-authors read the paper

extremely critically, word by word in the text, and line

by line in the Tables and captions, to check for errors.

Oftentimes, it is good also to have a trusted colleague

not directly involved in the work read your paper for

accuracy. The author who actually composes the paper

may not be able to spot his own errors or omissions. In

my experience, the paper with no inconsistencies or in-

accuracies in totals or other figures is a rarity.

I n c l u d e S e c ti o n o n W e a k n e s s e s o f Y o u r S t u d y

8. Be sure to include a section on weaknesses of the

current study. No study is immune from weaknesses, not

even the largest, best-planned, double-blind, random-ized, prospective multi-institutional trials. Most ordi-

nary laboratory or clinical papers have major

weaknesses. Patient number may be smaller than one

would want. Relevance may hindered by pre-selection

bias before a certain population reaches a tertiary cen-

ter. Follow-up may not be as long as one would wish.

The editors an d reviewers will respect your work mo re if

you show an awareness of the weaknesses of your study

and enumerate these for your readership. Also, such

frankness is an ethical responsibility, as readers may

alter their care of patients based on your recommenda-

tions. Being aware of the limitations of your work will

help the reader to determine if he feels your conclusions

are firmly enough established to warrant a change in his

clinical practice.

A v o i d S u p e r l a t iv e s i n D e s c r i b i n g y o u r W o r k

9. What I mean by this is that you should not succumb

to saying that your work is "the first" or the "biggest

T h e E d i t o r is' A l w a y s R i g h t

10. By and large, accept what your editor says and

wants. He was chosen for his experience and insight. He

is completely objective. He has enlisted expert opinions

regarding your work rendered by recognized authorities.

He directly controls whether or not your manuscript will

become a published paper. Do what he wants in modi-

fying your manuscript. Also--and this is very impor-tant-provide the editor a clear, concise tabulation of

the changes you have rendered in you r paper in response

to his and the reviewer's comments.

S t a t i s t i c a l C o n s u h a t i o n

11. Enlist statistical assistance liberally. Today's stan-

dards in terms of statistical analysis for leading jour-nals are very high. If major data analysis is inherent in

your research paper, it is often advisable to enlist the

support and advice of a trained statistician in analyzing

and interpreting your data. This can prevent later

problems in the review process and also enable you to

respond more meaningfully to any statistical issues that

are raised.

D o n o t b e d i s c o u r a g e d i f y o u r p a p e r i s r e j e c t e d

o n o n e o r m o r e s u b m i s s i o n s . fo r r e v i e w

12. Some o f the very best papers, which eventu ally come

to be respected and frequently quoted, are not accepted

on initial submission. You can learn from the reviewer's

comments and submit your work to another journal.

Anyone with significant experience in academic medi-

cine has learned to become inured to the rejection pro-cess. This does not necessarily mean that your work is

unworthy. You must persevere without getting dis-

couraged.The author hopes that these observations will be of

some use to the novice or occasional medical essayist

and may help in achieving acceptance of your manu-

scripts for publication.

12 Tips for Sci Paper

Documents