Scientific Writing

1
Scientific Writing

• October 8, 2008
• Jeanne Erdmann

2
Today

• Practical advice for good writing
• How you need to modify writing style
• Levels of editing
• Writing tips you can follow when you write a
  manuscript or grant proposal

3
Keep in Mind

• Your reviewers are busy, toothats why they call
  it peer review

4
Reviewers Workload

• 80 applications to review three times a year
• 16 personal assignments--proposals reviewer must
  read and be prepared to discuss
• 6 reviews to write (three-four pages each single
  spaced)
• 4 weeks to get it done
• 2 days away from home
• (Source ORS grant writing seminar)

5
Think About That Reviewer

• At the end of a long, busy day, your reviewer is
  sitting in a hot room fighting off sleep. He/she
  has already read five grants. Yours is next
• Beginning with the titlewhat kind of impression
  will you make?

6
Read Like a Writer

• Before you do anything, learn to protect your
  time. Preparing a grant or manuscript takes long,
  uninterrupted hours.
• Train yourself to read like a writer. You already
  read medical journalsexamine structure, writing
  style, overall organization, such as use of
  subheads.

7
While Youre Reading, Look For

• Clear, unambiguous sentences
• Coherent transitions
• Active rather than passive verbs
• Main points up front, stated clearly

8
Before Writing Your Own Manuscript

• Identify the unique element of your research. For
  example We can grow cartilage in the laboratory
  without a scaffold.
• Choose a focused hypothesis.
• From this moment on, every piece of information
  that you include should cling to the spine of
  this idea.
• Look on the bright side Youre already familiar
  with the process of research writing is a
  similar process.

9
Good Planning

• Think about the questions your readers or
  reviewers may have about your topic before you
  start writing and then answer them in your text.
  

10
Think, Focus, Then Write

• Resist the urge to get something down on paper.
  You will end up polishing paragraphs that
  shouldnt even be includedplanning saves editing
  time. (Even writers like to procrastinate first
  draft).
• Begin writing only when you can summarize your
  idea in one, irreducible sentence. (For example,
  scan the Research Articles in the TOC of
  Science).

11
Example - One Sentence Summary

• JAMA Vol. 300 No. 13, October 1, 2008
•  
• Title Variants of the Adiponectin (ADIPOQ) and
  Adiponectin Receptor 1 (ADIPOR1) Genes and
  Colorectal Cancer Risk
•  

12
JAMA Abstract

• First Sentence of that Abstract
• Current epidemiological evidence suggests an
  association between obesity, hyperinsulinemia,
  and colorectal cancer risk.

13
Titles Rule

• Give your manuscript a clear, explanatory
  title--A short, focused title will guide your
  writing focus.

14
Active Voice

• Active voice puts the focus on the scientists
• Here we present the previously uncharacterized
  structure of the ternary complex representing...

15
Why Active Voice?

• Uses fewer words
• Strongly links you to your work
• Forces you to find great verbs
• Keeps text lively

16
Passive voice

• Passive voice keeps the focus on the work, not
  the scientists doing the work
• To date, nearly 40 TGF-b family members have
  been isolated in the human genome, with which
  five type II and seven type I receptors
  interact.

17
Some Unnecessary Uses of Passive Voice

• Example Therefore, it is not unexpected that
  many physicians feel.
• Rewrite Therefore, many physicians feel

18
More Unnecessary Uses of Passive Voice

• Example It is anticipated that data collected
  from the proposed studies will
• Rewrite We anticipate that data
• Another Rewrite Data from the proposed studies
  should...

19
Passive Ex. Continued

• Example Consequently, it will be important to
  develop a strategy to manage the compromised
  annulus
• Rewrite Developing a strategy to manage the
  compromised annulus will be critical

20
Sound Confident

• Sometimes choosing the passive voice means
  choosing between confidence and clarity.

21
Overall Structure Needs Clarity

• Coherence reflects good transitions from sentence
  to sentence and from paragraph to paragraph. This
  means that each sentence in a paragraph flows
  into the next sentence and each paragraph flows
  into the next

22
Achieve Clarity Write Your Abstract First

• Clear, concise summary of your proposal (acts as
  your outline).
• Includes the major concepts that reflect the
  scope of your work.
• First place to show the strength of your idea
• First impression of your work
• The first sentence of your abstract should make
  the reviewer want to keep reading.

23
Good Paragraphs are a Well-Told Joke

• Think of a good joke. Each line sets up the
  next, until you reach the perfect punch line.
• In science-writing lingo, this means
• Structure each paragraph such that each sentence
  follows a logical order that builds to a strong,
  logical punch line.

24
Paragraphs, cont.

• The first sentence of your paragraphs signals
  what the rest of the paragraph is about. Gives
  you two ways to check for coherency.

25
Paragraphs, cont.

• Each sentence that follows should reflect back on
  that lead sentence.
• When youre editing for flow you can read the
  first sentence of each paragraph and see whether
  that tells a story.

26
Tips to Keep You Going

• Begin each session by reading your abstracthelps
  keep you focused.
• Once in a while, read out loud Read first
  sentence each paragraph. Read your subject
  headings.
• End each writing session in the middle of a
  sentence. Begin the next session by completing
  that sentence. (Sounds dumb but it gets your
  fingers moving and it works!)

27
Good Writing Means Careful Editing

• Edit for mistakes (grammar, spelling, mislabeled
  figures, numbers, and visual mistakes such as
  line spacing).
• Edit for clarity.
• Edit for overall appearance.

28
Edit for Clarity

• Define abbreviations.
• Use transitions from paragraph to paragraph and
  section to section. (An easy trick for clear
  transitions Repeat a word from the previous
  sentence).
• Vary sentence length.
• Rewrite ambiguous (vague, verbose) sentences.

29
Rewrite Vague, Verbose Sentences

• Vague   We expect that this service will result
  in marked improvements in patient outcomes.
• Clear This service will prevent peri-operative
  pulmonary emboli.
• (Example from Dr. Gage)

30
Rewrite Ambiguous Sentences

• Vague  Prior to leaving the hospital, the nurse
  will review each prescribed medication. sounds
  like the nurse is leaving
• Clear Prior to patient discharge, the floor
  nurse will review each prescribed medication.
• (Example from Dr. Gage)

31
Use a Single Thought per Sentence

• Overstuffed Young cartilage cells, which provide
  the Companys proprietary restorative process,
  possess unique immune-privileged status with a
  naturally enhanced capacity to generate new
  cartilage.

32
Rewritten as Two Sentences

• Chondrocytes derived from young donors regenerate
  cartilaginous grafts 100-fold more efficiently
  than adult cartilage cells.
• Because chondrocytes are immune privileged, the
  Companys laboratory-grown cartilage can be
  transplanted without fear of rejection.

33
Another example

• InQu Bone Graft Extender and Substitute
  represents a new approach to optimizing bone
  healing that combines the structural and
  biological attributes inherent to two
  biomaterials with a long history of safe and
  effective clinical use, and which that offer a
  number of advantages to the surgeon.

34
One RewriteMade Two Grafs

• InQu Bone Graft Extender and Substitute joins
  the structural and biological attributes of PLGA
  and Hyaluronic acid, two biomaterials with a long
  history of safe and effective clinical use. This
  fusion offers a product that optimizes bone
  healing by creating a unique microenvironment
  that is both biocompatible and conducive to new
  bone formation. The three-dimensional structure
  based on the resorbable properties of PLGA
  supports safe and reliable bone resorption at the
  site of implantation. Hyaluronic acid, a
  molecule critical for tissue regeneration and
  repair, aids essential early events in bone
  formation by promoting the migration,
  proliferation, and differentiation of osteogenic
  cells.

35
Edit for Appearance

• Consistent font
• Consistent line spacing
• Break up text with bullets
• Include figures, drawings, photos
• Use subheads
• (White space makes a manuscripts easier to read.)

36
Cut Unnecessary Words

• Consensus not- Consensus of opinion
• To - not -In order to
• Despite-not-Despite the fact that
• (Refer to handout for additional examples)

37
More Unnecessary Words

• Eliminate prepositional phrases Prepositional
  phrases often waste space and bog down sentences.
  Cut them whenever a shorter formworks better.
• Ex Replace "the purpose of the study" with "the
  study's purpose.".

38
Sample Sentence Unnecessary Introductory Clause

• One hypothesis to explain late-onset
  aggregation-associated toxicity posits that the
  deposition of toxic aggregates is a stochastic
  process, governed by a nucleated polymerization
  and requiring many years to initiate disease.
• Came from 2006 Science article

39
Just Say

• The deposition of toxic aggregates, a stochastic
  process, could be governed by nucleated
  polymerization that begins many years before
  disease onset.

40
Do Your Numbers Add Up?

• Have you accounted for all of your study
  participants? (Do number of patients in treatment
  and control groups match total number of
  participants?)
• Do your percentages add up to 100?
• Do your percentages and raw numbers agree?
• (Source BMJ Technical Editing for Copyeditors
  2005 http//resources.bmj.com/bmj/interactive/talk
  s)

41
Checklist

• The Title and Abstract reflect the focus of the
  work you are proposing.
• Your manuscript flows from sentence to sentence,
  from paragraph to paragraph, from section to
  section, from beginning to end.
• Youve broken up text with photos, illustrations,
  bullets, creating plenty of white space.
• Youve kept the document within recommended page
  limit or word count.

42
Common Problems I Encounter

• Absolutely no context--Failing to explain the
  significance of the work.
• Subjects and verbs separated by a cascade of
  incoherent clauses.
• Burying your best, strongest, point under a
  cascade of incoherent clauses.
• Too much jargon thats not well defined.

43
What The?

• Over the past decades, the application of
  dynamic modeling and control has aided
  understanding of the complexities of their
  dynamics and control.
• Ms from the International Federation of Automatic
  Control Seoul Korea July 6-11, 2008)

44
Grant Writing Tips

• Put your proposal into the fundable pile.

45
Begin 3 Months Before the Deadline

• Might find need more data
• Need time make certain idea original
• Need time provide detailed methodology
• Budget, face sheets, letters from collaborators
  time-consuming to assemble
• Allows time for a cold reading.
• Nancy Shinowara a SRO at the NIH said this to me
  "Its a mistake not to spend a lot of time on
  grants. Sloppy work really ticks off reviewers.

46
What Makes a Strong Proposal?

• A drop-dead, great idea. (Same requirement for me
  to sell a news feature to Science, Nature, or
  Selfmy editor at Science told me recently that
  ideas need to be blockbuster)
• Information that convinces the study section you
  are capable of following through.
• A clearly written and well organized document.

47
How Can You succeed in this Environment?

• Work very closely with a mentor. He/she should
  read and critique your grant from the top down.
  For example, check for holes in the science, gaps
  in technical feasibility.
• If your proposal includes a well-known scientist
  as a collaborator, make sure that your work is
  not shoddy (the committee will wonder whether
  this scientist is truly closely involved or
  whether youve just used his/her name to
  improve your chances of funding).
• Many institutions that successfully fund
  first-time applicants form committees to critique
  proposals.
• Source Mehrdad Tondravi Ph.D. Scientific Review
  Administrator, NKKD, Division of Physiology and
  Pathology

48
Criticism is not Personal

• At this point, you dont want praise. You want
  substantive feedback.
• Your study section will be far more critical.

49
Tips for Clinical Research Proposals

• Make certain that your biostatistics have enough
  power to support or refute your hypothesis.
• (More than half of the proposals submitted to
  NIDDK fail because of inadequate statistics)
• Show that either you or your collaborator has a
  clinical history of sufficient recruitment and
  retention.
• (Source Mehrdad Tondravi SRA Skeletal Biology
  Structure and RegenerationOct 2007)

50
Red Flags for Clinical Research Proposals

• Insufficient numbers to refute or uphold your
  hypothesis
• Power calculations done incorrectly
• Demographics prevalence of proposed disease in
  your area too low to meet inclusion criteria
  (This could be a problem if you claim that youll
  have a 95 retention rate).
• Really need to demonstrate that you can recruit
  and retain a sufficient number of subjects.
• (Source Mehrdad Tondravi SRA Skeletal Biology
  Structure and RegenerationOct 2007)
• (Source Mehrdad Tondravi SRA Skeletal Biology
  Structure and Regeneration)

51
Tips continued

• Make certain your approach is appropriate for the
  questions asked.
• Describe how you will retain research subjects,
  especially if study is long term.
• Make certain that your criteria for inclusion and
  exclusion are medically sound.

52
Remember Tell a Good Story

• Never forget context Every section of your
  proposal should mention the relevance of your
  idea.
• The Research Plan is not a review article. Its
  where you show the timeliness of your idea by
  answering, Why now? In your discussion of the
  literature, always refer back to your project
  idea and say why its relevant.

53
Tone and Appearance

• To make your grant readable and understandable
  here's some advice from Anthony M. Coelho, Ph.D.,
  NIH
• For organization think USA Today clear,
  readable, text nice little chunks, lots of
  pictures.
• For tone think Scientific American.
• Making your reviewer's job easy is not the same
  as dumbing down.

54
Think Perfection

• Youll never achieve perfection but your your
  reviewers will be happy.

55
Hints on Letters of Support

• Don't submit letter from collaborator or
  supporter with one or two lame sentences ("I have
  great enthusiasm for this proposal").
• Submit one rich in detail. What will your
  collaborators do? When? Where? How?
• Write this letter yourself. Tell your
  collaborator that you know he/shes busy and your
  "helping" them out by providing a draft letter.

56
Avoid Common Problems

• Again, follow the directions--page length, margin
  size, and the order in which you put everything
  together.
• Watch page length (Its not OK to put preliminary
  data into an appendix).
• Make certain you are not proposing too much work.
  
• Write a reasoned, critical research plan that
  includes the pitfalls of your approach and how
  you will handle them. Better for you to address
  potential problem areas and how you will deal
  with them than leave reviewers wondering.
• (Source NIH Writing Workshops)

57
Get to Know the Staff at Funding Agencies

• Call them discuss your idea. The staff can help
  you decide whether your project is ready for
  funding. When in doubtask.
• Once agency staff knows you, your phone calls are
  returned more quickly.
• Attend a grant writing seminar. Bring your
  specific aims page get some feedback.

58
If At First You Dont Succeed

• If you need to write a revision
• Read your critique carefully and pay close
  attention to reviewers comments.
• If a reviewer seems to have missed the point,
  blame yourself for not making the writing clear
  enough. For example, Needs collaboration with
  immunologist
• When submitting your revision, make certain that
  youve addressed every criticism.

59
Final Reminder

• Dont ever be tempted to turn in a
  less-than-stellar effort. Study sections have
  long memories and you will get a reputation that
  will be difficult to turn around.
• Don't run everything to the last minute but just
  in case you doFedEx has same-day service to the
  NIH, more than 200 you need to schedule before
  6 a.m.

60
Where to Get Help

• Books
• Bugs in Writing A Guide to Debugging Prose by
  Lynn Dupre
• A Writers Reference by Diana Hacker
• Barrass, Robert. (2002). Scientists Must Write,
  2nd edition. New York Routledge.

61
Websites for Grant Writing

• NIH
• http//deainfo.nci.nih.gov/EXTRA/EXTDOCS/gntapp.h
  tm
• http//www.niaid.nih.gov/ncn/grants/write/write_pf
  .htm
• NIH Grant Writing Tips Sheets http//grants1.nih.
  gov/grants/grant_tips.htm
•  
• Science
• http//sciencecareers.sciencemag.org/career_devel
  opment/previous_issues/articles/0210/grantdoctor_i
  ndex_of_columns
• http//nextwave.sciencemag.org/cgi/content/full/1
  999/09/20/2
• BioScienceWriters
• http//www.biosciencewriters.com/pages/GrantTips.
  htm
•  

62
Blog For Scientific Writing

• Time for a change, by Linda Cooper on the
  Nature Network (www.Nature.com)
• http//network.nature.com/blogs/user/UA8E0D68F

63
Journal Websites

• BMJ has a great site
• http//resources.bmj.com/bmj/interactive/talks

64
Handouts on Class Website

• The Science of Scientific Writing
• George D. Gopen and Judith A. Swan. This Article
  appeared in the American Scientist (Nov-Dec
  1990), Volume 78, 550-558.
• Cut Unnecessary Words

65
Repetitive Advice?

• Finally, you read handouts and blog youll notice
  that advice may begin to sound repetitivethats
  because there are tried and true formulas for
  manuscript writing and writing in genera. No
  magic bullet, just hard, hard, work.

66
Jeanne Erdmannerdmannj_at_nasw.org