Successful Scientific Writing

1
Successful Scientific Writing
Paul Z. Siegel, M.D., M.P.H. Based on
materials developed in collaboration with Richard
A. Goodman, M.D., J.D.
2

• in close proximity to - near
• the predominant number of
• in a large number of cases
• on a previous occasion
• in the absence of
• with regard to
• at some future time
• due to the fact that
• serves the function of being

3

• heretofore

4

• heretofore - until now
• theretofore - before then
• remuneration - payment
• promulgate - to publish
• - make widely known

5

• I am a bear of
• very little brain
• and long words bother me.

6
A Dictionary of Useful Research Phrases

• From Thurlbeck, WM. Glasnost and communication.
• Am Rev Respir Dis 1989139282-283.

7
A Dictionary of Useful Research Phrases

• "It has long been known..."

8
A Dictionary of Useful Research Phrases

• "It has long been known..."
• I didn't look up the original references.

9
A Dictionary of Useful Research Phrases

• "It has long been known..."
• I didn't look up the original references.
• "It is believed that..."

10
A Dictionary of Useful Research Phrases

• "It has long been known..."
• I didn't look up the original references.
• "It is believed that..."
• I think.

11
A Dictionary of Useful Research Phrases

• "It has long been known..."
• I didn't look up the original references.
• "It is believed that..."
• I think.
• "It is generally believed that..."

12
A Dictionary of Useful Research Phrases

• "It has long been known..."
• I didn't look up the original references.
• "It is believed that..."
• I think.
• "It is generally believed that..."
• My friends think so too.

13
A Dictionary of Useful Research Phrases

• "It has long been known..."
• I didn't look up the original references.
• "It is believed that..."
• I think.
• "It is generally believed that..."
• My friends think so too.
• "Correct within an order of magnitude..."

14
A Dictionary of Useful Research Phrases

• "It has long been known..."
• I didn't look up the original references.
• "It is believed that..."
• I think.
• "It is generally believed that..."
• My friends think so too.
• "Correct within an order of magnitude..."
• Wrong.

15
A Dictionary of Useful Research Phrases

• "It has long been known..."
• I didn't look up the original references.
• "It is believed that..."
• I think.
• "It is generally believed that..."
• My friends think so too.
• "Correct within an order of magnitude..."
• Wrong.
• "A statistically oriented projection..."

16
A Dictionary of Useful Research Phrases

• "It has long been known..."
• I didn't look up the original references.
• "It is believed that..."
• I think.
• "It is generally believed that..."
• My friends think so too.
• "Correct within an order of magnitude..."
• Wrong.
• "A statistically oriented projection..."
• Wild guess.

17
PURPOSE

• Communicate more effectively

18
PURPOSE

• Communicate more effectively
• Publish more efficiently

19
(No Transcript)
20

• Communicate effectively

21
To assess reliability of the questionnaire, a
test-retest study was conducted. Agreement was
high for sociodemographic variables. Reliability
of information on chronic conditions was also
high.

• Do reliability and agreement mean the same
  thing?

22
To assess reliability of the questionnaire, a
test-retest study was conducted. Agreement
Reliability was high for sociodemographic
variables. Reliability of information on chronic
conditions was also high.

23
To assess reliability of the questionnaire, a
test-retest study was conducted. Reliability was
high for sociodemographic variables and chronic
conditions.

• Clear
• Smooth

24
?
?
reliability agreement
reliability
25
?
?
reliability agreement
reliability

• of thought

26
If heterosexual transmission was, in fact, the
mode of HIV infection for persons who reported
other risks as well as heterosexual contact, for
persons who are currently reported without risks
but who may be reclassified in the future to the
heterosexual-contact category, as well as for
persons formerly classified as "Pattern II
heterosexually acquired," then 12 of cumulative
and 16 of cases reported in 1993 may be
attributable to heterosexual transmission.
?
or
27
If heterosexual transmission was, in fact, the
mode of HIV infection for persons who reported
other risks as well as heterosexual contact, for
persons who are currently reported without risks
but who may be reclassified in the future to the
heterosexual-contact category, as well as for
persons formerly classified as "Pattern II
heterosexually acquired," then 12 of cumulative
and 16 of cases reported in 1993 may be
attributable to heterosexual transmission.

• Strain factors
• long sentences (gt25 words)
• conjecture

28
If heterosexual transmission was, in fact, the
mode of HIV infection for persons who reported
other risks as well as heterosexual contact, for
persons who are currently reported without risks
but who may be reclassified in the future to the
heterosexual-contact category, as well as for
persons formerly classified as "Pattern II
heterosexually acquired," then 12 of cumulative
and 16 of cases reported in 1993 may be
attributable to heterosexual transmission.
Conjecture
29
SECTIONS OF A SCIENTIFIC ARTICLE

• 2. ABSTRACT
• 3. INTRODUCTION
• 4. METHODS
• 5. RESULTS
• 6. DISCUSSION

30
SECTIONS OF A SCIENTIFIC ARTICLE

• 1. TITLE
• 2. ABSTRACT
• 3. INTRODUCTION
• 4. METHODS
• 5. RESULTS
• 6. DISCUSSION

31
The ABSTRACT
Purpose to highlight key points from the major
sections of the article
Component of abstract Abstracted from Component of abstract Abstracted from
? Major purpose of study ? Basic procedures ? Main findings ? Principal conclusions Methods Results Discussion  
32
The ABSTRACT
Purpose to highlight key points from the major
sections of the article
Component of abstract Abstracted from Component of abstract Abstracted from
? Major purpose of study ? Basic procedures ? Main findings ? Principal conclusions Introduction Methods Results Discussion  
33
The ABSTRACT
Purpose to highlight key points from the major
sections of the article
Component of abstract Abstracted from Component of abstract Abstracted from
? Major purpose of study ? Basic procedures ? Main findings ? Principal conclusions Introduction Methods Results Discussion  

• List Medical Subject Headings (MeSH) key words
  under the abstract
• Emphasize what is new and useful.

34

• Emphasize what is
• new and useful.

35
Screening research papers by reading abstracts

• Please get the abstract right, because we may
  use it alone to assess your paper.
• -- British Medical Journal

36
Reliability of Information on Chronic Disease
Risk Factors Collected in the Missouri Behavioral
Risk Factor Surveillance System Abstract
(initial draft - 183 words) small-group
exercise
37
Background Info

• BRFSS what is it?
• Validity vs. Reliability
• Kappa values

38
Editing

• SUBSTANCE
• Mistakes
• Ambiguity
• Missing key info
• Extraneous/misleading info
• Empty statements
• STYLE

39
PURPOSE

• The Behavioral Risk Factor Surveillance System
  (BRFSS) is widely used by state health agencies
  to measure the prevalence of chronic disease risk
  factors. Despite the widespread use of BRFSS,
  few studies exist on the reliability and validity
  of BRFSS-collected data. To assess the
  reliability of the Missouri BRFSS . . .

40
The Behavioral Risk Factor Surveillance System
(BRFSS) is widely used by state health agencies
to measure the prevalence of chronic disease risk
factors. Despite the widespread use of BRFSS,
few studies exist on the reliability and
validity of BRFSS-collected data. To assess the
reliability of the Missouri BRFSS . . .

• The Behavioral Risk Factor Surveillance System
  (BRFSS) is a population-based telephone survey of
  health-related behaviors among adults ages 18 and
  older that is used by nearly all state health
  agencies to measure the prevalence of chronic
  disease risk factors. Despite widespread use,
  the reliability of BRFSS-collected data is not
  well described (understood . . . characterized .
  . . ). To assess the reliability of the Missouri
  BRFSS . . .

41
METHODS

• . . . , a test-retest study
  was conducted. The authors conducted telephone
  reinterviews for 222 respondents of completed
  BRFSS interviews from March and April 1993. The
  second interview was completed between six and 30
  days of the first interview.

42
PASSIVE VOICE

• Expresses action without specifying who or what
  performed the action
• (the agent of the action is omitted)
• "James Watson was awarded the Nobel Prize for
  discovering the molecular structure of DNA.

43
PASSIVE VOICE

• Expresses action without specifying who or what
  performed the action
• (the agent of the action is omitted)
• "James Watson was awarded the Nobel Prize for
  discovering the molecular structure of DNA.
• vs.
• "The Nobel Committee awarded James Watson the
  Nobel Prize for discovering the molecular
  structure of DNA."

44
A Controlled Trial of Web-Based Diabetes
Disease Management Meigs, JB et al. A
Controlled Trial of Web-Based Diabetes Disease
Management. Diabetes Care 26750-757, 2003

• One group of physicians (intervention group) was
  trained to use a web-based information
  management/ clinical decision support tool. The
  other group of physicians (control group) was
  not.
• The study was a group randomized, controlled
  trial. A coin was tossed to select an
  intervention group and a control group.

45
Preventing Pneumococcal Disease Among Infants and
Young Children

• All children aged lt23 months should be
  vaccinated with PCV7.
• Diffuse the locus of responsibility

46
METHODS

• . . . , a test-retest study
  was conducted. The authors conducted Telephone
  reinterviews were conducted for 222 respondents
  of completed BRFSS interviews from March and
  April 1993. The second interview was completed
  between six and 30 days of the first interview.

47
METHODS

• . . . , a test-retest study
  was conducted. Telephone reinterviews were
  conducted for 222 respondents of completed BRFSS
  interviews from March and April 1993. The second
  interview was completed between six and 30 days
  of the first interview.

48
METHODS

• . . . , a test-retest study
  was conducted. Interviews from 222 of 252
  respondents (response rate 88) who completed
  the survey during March and April, 1993, were
  repeated between six and 30 days after the
  original interview.

49
METHODS

• . . . , a test-retest study
  was conducted. Interviews from 222 of 252
  respondents (response rate 88) who completed
  the survey during March and April, 1993, were
  repeated between six and 30 days after the
  original interview.

What if your abstract has 210 words and the
maximum is 200? Can we make this any shorter?
50
. . . , a test-retest
study was conducted interviews from 222 of 252
respondents (response rate 88) who completed
the survey during March and April, 1993, were
repeated between six and 30 days after the
original interview.

• . . . interviews from 222 of 252 respondents who
  completed the survey during March and April,
  1993, were repeated 6-30 days after the original
  interview.
• 34 words ? 23 words

51
RESULTS

• Agreement was high for sociodemographic variables
  (kappa values from 0.85 to 1.00). Reliability of
  information on chronic conditions and risk
  factors was also high, with kappa values from
  0.82 for hypertension to 1.00 for current smoking
  status.
• Regarding cancer screening practices, reliability
  was lower for knowledge of the prostate-specific
  antigen test (kappa 0.21) than for women's
  cancer screening practices (i.e., the mammogram
  and Pap smear).
• Questions on attitudes toward environmental
  tobacco smoke showed lower reliability than did
  questions on individual actions to reduce
  exposure to environmental tobacco

52
RESULTS

• Agreement was high for sociodemographic variables
  (kappa values from 0.85 to 1.00). Reliability of
  information on chronic conditions and risk
  factors was also high, with kappa values from
  0.82 for hypertension to 1.00 for current smoking
  status.
• Regarding cancer screening practices, reliability
  was lower for knowledge of the prostate-specific
  antigen test (kappa 0.21) than for women's
  cancer screening practices (i.e., the mammogram
  and Pap smear).
• Questions on attitudes toward environmental
  tobacco smoke showed lower reliability than did
  questions on individual actions to reduce
  exposure to environmental tobacco

53
RESULTS

• Agreement was high for sociodemographic variables
  (kappa values from 0.85 to 1.00). Reliability of
  information on chronic conditions and risk
  factors was also high, with kappa values from
  0.82 for hypertension to 1.00 for current smoking
  status.
• Regarding cancer screening practices, reliability
  was lower for knowledge of the prostate-specific
  antigen test (kappa 0.21) than for women's
  cancer screening practices (i.e., the mammogram
  and Pap smear).
• Questions on attitudes toward environmental
  tobacco smoke showed lower reliability than did
  questions on individual actions to reduce
  exposure to environmental tobacco

54

• Agreement was high for sociodemographic variables
  (kappa values from 0.85 to 1.00). Reliability of
  information on chronic conditions and risk
  factors was also high, with kappa values from
  0.82 for hypertension to 1.00 for current smoking
  status.

55

• Regarding cancer screening practices, reliability
  was lower for knowledge of the prostate-specific
  antigen test (kappa 0.21) than for women's
  cancer screening practices (i.e., the mammogram
  and Pap smear).

56

• Questions on attitudes toward environmental
  tobacco smoke (kappa 0.47- 0.51) showed lower
  reliability than did questions on individual
  actions to reduce exposure to environmental
  tobacco smoke (kappa 0.62 - 0.82).

57
Reliability was high for sociodemographic
variables, chronic conditions and risk factors
(kappa gt 0.82).Reliability was lower for
knowledge of the prostate-specific antigen test
(kappa 0.21) than for women's cancer screening
practices (kappa 0.59 - 0.87). Questions on
attitudes toward environmental tobacco smoke
(kappa 0.47- 0.51) showed lower reliability
than did questions on individual actions to
reduce exposure to environmental tobacco smoke
(kappa 0.62 - 0.82).
58
Reliability was high for sociodemographic
variables, chronic conditions and risk factors
(kappa gt 0.82).Reliability was lower for
knowledge of the prostate-specific antigen test
(kappa 0.21) than for women's cancer screening
practices (kappa 0.59 - 0.87). Questions on
attitudes toward environmental tobacco smoke
(kappa 0.47- 0.51) showed lower reliability
than did questions on individual actions to
reduce exposure to environmental tobacco smoke
(kappa 0.62 - 0.82).
59
Reliability was high for sociodemographic
variables, chronic conditions and risk factors
(kappa gt 0.82), lower for cancer screening
practices (kappa 0.59-0.87), and lowest for
knowledge about prostate-specific antigen testing
(kappa 0.21). Reliability of questions about
attitudes toward environmental tobacco smoke
(kappa 0.47-0.51) was lower than for questions
about individual actions to reduce environmental
tobacco smoke (kappa 0.62-0.82).
60

• one final detail

61
Reliability was highest for sociodemographic
variables, chronic conditions and risk factors
(kappa gt 0.82), lower for cancer screening
practices (kappa 0.59-0.87), and lowest for
knowledge about prostate-specific antigen testing
(kappa 0.21). Reliability of questions about
attitudes toward environmental tobacco smoke
(kappa 0.47-0.51) was lower than for questions
about individual actions to reduce environmental
tobacco smoke (kappa 0.62-0.82).
62
RESULTS

• Original 87 words
• Revision 65 words (and more information)

63
CONCLUSION

• These findings demonstrate the overall
  flexibility and utility of the BRFSS.
• ?
• We conclude that the reliability of most BRFSS
  questions is high. New, more reliable, questions
  to measure knowledge of prostate-specific antigen
  testing need to be developed.

64
The Abstract

• RELIABILITY
• VALIDITY

• Purpose

65
The Abstract

• RELIABILITY
• VALIDITY
• RELIABILITY

• Purpose
• Methods

66
The Abstract

• RELIABILITY
• VALIDITY
• RELIABILITY
• RELIABILITY

• Purpose
• Methods
• Results

67
The Abstract

• RELIABILITY
• VALIDITY
• RELIABILITY
• RELIABILITY
• FLEXIBILITY
• UTILITY

• Purpose
• Methods
• Results
• Conclusions

68
The Abstract

• RELIABILITY
• VALIDITY
• RELIABILITY
• RELIABILITY
• FLEXIBILITY
• UTILITY

• Purpose
• Methods
• Results
• Conclusions

69
The Behavioral Risk Factor Surveillance System
(BRFSS), a population-based telephone survey of
health-related behaviors among adults ages 18 and
older, is used by nearly all state health
agencies to measure the prevalence of chronic
disease risk factors. Despite widespread use,
the reliability of BRFSS-collected data is not
well described. To assess the reliability of the
Missouri BRFSS, interviews from 222 of 252
respondents who completed the survey during March
and April, 1993, were repeated 6-30 days after
the original interview. Reliability was highest
for sociodemographic variables, chronic
conditions and risk factors (kappa gt 0.82), lower
for cancer screening practices (kappa
0.59-0.87), and lowest for knowledge about
prostate-specific antigen testing (kappa 0.21).
Reliability of questions about attitudes toward
environmental tobacco smoke (kappa 0.47-0.51)
was lower than for questions about individual
actions to reduce environmental tobacco smoke
(kappa 0.62-0.82). We conclude that the
reliability of most BRFSS questions is high.
New, more reliable, questions to measure
knowledge of prostate-specific antigen testing
need to be developed. (165 words)
summary of results recommendation
70
The Behavioral Risk Factor Surveillance System
(BRFSS), a population-based telephone survey of
health-related behaviors among adults ages 18 and
older, is used by nearly all state health
agencies to measure the prevalence of chronic
disease risk factors. Despite widespread use,
the reliability of BRFSS-collected data is not
well described. To assess the reliability of the
Missouri BRFSS, we conducted a test-retest study
among 222 (88) of 252 respondents who completed
the survey during March and April, 1993. We also
tested the flexibility and utility of the BRFSS
by adding questions to the core questionnaire
about use/knowledge of prostate specific antigen
screening and exposure to environmental tobacco
smoke. Reliability was highest for
sociodemographic variables, chronic conditions
and risk factors (kappa gt 0.82), lower for cancer
screening practices (kappa 0.59-0.87), and
lowest for knowledge about prostate-specific
antigen testing (kappa 0.21). Reliability of
questions about attitudes toward environmental
tobacco smoke (kappa 0.47-0.51) was lower than
for questions about individual actions to reduce
environmental tobacco smoke (kappa 0.62- 0.82).
These findings demonstrate the flexibility and
utility of the BRFSS as well as the high
reliability of most BRFSS questions. New, more
reliable, questions to measure knowledge of
prostate-specific antigen testing need to be
developed. (199 words)
71
Structured Abstracts
72
Structured Abstracts
Context  Objective  Design and
Setting Patients  Main Outcome Measure  Results
Conclusions  Background Methods Results Conclus
ions
73
Structured Abstracts
Context  Objective  Design and
Setting Patients  Main Outcome Measure  Results
Conclusions  Background Methods Results Conclus
ions
74
Any Volunteers?
75
The INTRODUCTION
76
The INTRODUCTION
Purpose to convince the reader that your study
will yield knowledge or knowhow that is new and
useful
77
The INTRODUCTION
Purpose to convince the reader that your study
will yield knowledge or knowhow that is new and
useful
78
The INTRODUCTION
Purpose to convince the reader that your study
will yield knowledge or knowhow that is new and
useful
Components of the Introduction
 
79
The INTRODUCTION
Purpose to convince the reader that your study
will yield knowledge or knowhow that is new and
useful
Components of the Introduction
? Identify a gap in knowledge or knowhow (Study problem).  
80
The INTRODUCTION
Purpose to convince the reader that your study
will yield knowledge or knowhow that is new and
useful
Components of the Introduction
? Identify a gap in knowledge or knowhow (Study problem). - Provide key background (scope/nature/magnitude of the gap).  
81
The INTRODUCTION
Purpose to convince the reader that your study
will yield knowledge or knowhow that is new and
useful
Components of the Introduction
? Identify a gap in knowledge or knowhow (Study problem). - Provide key background (scope/nature/magnitude of the gap). - Be clear that filling this gap will be useful.  
82
The INTRODUCTION
Purpose to convince the reader that your study
will yield knowledge or knowhow that is new and
useful
Components of the Introduction
? Identify a gap in knowledge or knowhow (Study problem). - Provide key background (scope/nature/magnitude of the gap). - Be clear that filling this gap will be useful. - Describe the relevant limitations of previous studies.  
83
The INTRODUCTION
Purpose to convince the reader that your study
will yield knowledge or knowhow that is new and
useful
Components of the Introduction
? Identify a gap in knowledge or knowhow (Study problem). - Provide key background (scope/nature/magnitude of the gap). - Be clear that filling this gap will be useful. - Describe the relevant limitations of previous studies. ? Present your approach to filling the gap (Study purpose).  
84
The INTRODUCTION
Purpose to convince the reader that your study
will yield knowledge or knowhow that is new and
useful
Components of the Introduction
? Identify a gap in knowledge or knowhow (Study problem). - Provide key background (scope/nature/magnitude of the gap). - Be clear that filling this gap will be useful. - Describe the relevant limitations of previous studies. ? Present your approach to filling the gap (Study purpose). - Be clear that your approach is new.  
85
The INTRODUCTION
Purpose to convince the reader that your study
will yield knowledge or knowhow that is new and
useful
Components of the Introduction
? Identify a gap in knowledge or knowhow (Study problem). - Provide key background (scope/nature/magnitude of the gap). - Be clear that filling this gap will be useful. - Describe the relevant limitations of previous studies. ? Present your approach to filling the gap (Study purpose). - Be clear that your approach is new. - Emphasize that your approach addresses the limitations of previous studies in a logical/compelling way.  
86
The INTRODUCTION
Purpose to convince the reader that your study
will yield knowledge or knowhow that is new and
useful
Components of the Introduction
? Identify a gap in knowledge or knowhow (Study problem). - Provide key background (scope/nature/magnitude of the gap). - Be clear that filling this gap will be useful. - Describe the relevant limitations of previous studies. ? Present your approach to filling the gap (Study purpose). - Be clear that your approach is new. - Emphasize that your approach addresses the limitations of previous studies in a logical/compelling way.  
87
The INTRODUCTION
Purpose to convince the reader that your study
will yield knowledge or knowhow that is new and
useful
Components of the Introduction
? Identify a gap in knowledge or knowhow (Study problem). - Provide key background (scope/nature/magnitude of the gap). - Be clear that filling this gap will be useful. - Describe the relevant limitations of previous studies. ? Present your approach to filling the gap (Study purpose). - Be clear that your approach is new. - Emphasize that your approach addresses the limitations of previous studies in a logical/compelling way.  

• (often requires just three paragraphs)

88
Elements of the Introduction

• GAP
• GB Key BACKGROUND to the gap.
• GU Why filling the gap will be
  USEFUL.
• GL Key LIMITATION(S) of previous
  studies responsible for the gap.
• APPROACH
• AB BACKGROUND to the approach (not
  always necessary)
• AN What is NEW about the approach
  taken in the current study.
• AL How does this new approach address
  the LIMITATIONS of previous
• studies.

89

• GU If radiofrequency exposure from mobile phone
  use increases the risk of cancer, acoustic
  neuroma would be of potential concern. GB
  The exposure from mobile phones is concentrated
  in the head close to the handset exposure is
  relatively high only for the glial and meningeal
  tissue closest to the surface of the head, the
  parotid gland, and the vestibular portion of the
  eighth cranial nerve where acoustic neuromas
  arise. Gap Six studies have investigated
  the association between mobile phone use and
  acoustic neuroma, with inconsistent results.
  GL All available studies are limited by a small
  number of exposed cases and a short follow-up
  since the time hand-held mobile phones first
  became available.

90

• GU If radiofrequency exposure from mobile phone
  use increases the risk of cancer, acoustic
  neuroma would be of potential concern. GB
  The exposure from mobile phones is concentrated
  in the head close to the handset exposure is
  relatively high only for the glial and meningeal
  tissue closest to the surface of the head, the
  parotid gland, and the vestibular portion of the
  eighth cranial nerve where acoustic neuromas
  arise. Gap Six studies have investigated
  the association between mobile phone use and
  acoustic neuroma, with inconsistent results.
  GL All available studies are limited by a small
  number of exposed cases and a short follow-up
  since the time hand-held mobile phones first
  became available.
• AB In Sweden, mobile phone use became common in
  the general population relatively early handheld
  mobile phones were introduced at the end of the
  1980s with an exponential increase of users
  during the 1990s. Slightly less than 6 of the
  population used mobile phones in 1990, 23 in
  1995, and over 80 today. Therefore, Approach
  a study based on the Swedish population (is this
  new?) AL will have a large proportion of
  long-term users, which is crucial for the
  possibility of detecting any increased risk of
  tumors related to long-term mobile phone use.

91

• GU If radiofrequency exposure from mobile phone
  use increases the risk of cancer, acoustic
  neuroma would be of potential concern. GB
  The exposure from mobile phones is concentrated
  in the head close to the handset exposure is
  relatively high only for the glial and meningeal
  tissue closest to the surface of the head, the
  parotid gland, and the vestibular portion of the
  eighth cranial nerve where acoustic neuromas
  arise. Gap Six studies have investigated
  the association between mobile phone use and
  acoustic neuroma, with inconsistent results.
  GL All available studies are limited by a small
  number of exposed cases and a short follow-up
  since the time hand-held mobile phones first
  became available.
• AB In Sweden, mobile phone use became common in
  the general population relatively early handheld
  mobile phones were introduced at the end of the
  1980s with an exponential increase of users
  during the 1990s. Slightly less than 6 of the
  population used mobile phones in 1990, 23 in
  1995, and over 80 today. Therefore, Approach
  a study based on the Swedish population AL
  will have a large proportion of long-term users,
  which is crucial for the possibility of detecting
  any increased risk of tumors related to long-term
  mobile phone use.

92
Lung Cancer Risk and Workplace Exposure
toEnvironmental Tobacco Smoke

• Group Exercise
• What is the Gap (G)?
• Does the study approach address the limitations
  of previous studies in a logical/compelling way?
• Why will filling this gap be useful (Gu)?

93
What is the Gap (G)?

• Exposure to environmental tobacco smoke (ETS)
  has been recognized as a cause of human cancer by
  the US Surgeon General,1 the National Institute
  for Occupational Safety and Health,2 the US
  Environmental Protection Agency,3 the California
  Environmental Protection Agency,4 the National
  Health and Medical Research Council of
  Australia,5 the Great Britain Department of
  Health,6 and most recently, the International
  Agency for Research on Cancer.7 Evidence for this
  association has come primarily from studies of
  nonsmokers who are married to a smoker, and
  meta-analyses of these studies have demonstrated
  strong and consistent evidence for an
  association.3,8,9
• Demonstrating an association between workplace
  ETS exposure and lung cancer risk has been more
  difficult. Early meta-analyses failed to
  demonstrate an association between workplace ETS
  exposure and lung cancer risk among
  nonsmokers,1014 but a statistically significant
  association has been reported in the 3 most
  recently published meta-analyses.1517 We sought
  to extend the previous meta-analyses by including
  additional studies and by conducting analyses
  stratified by level of exposure, which was not
  performed in the previous meta-analyses.

94
What is the Gap (G)?

• Exposure to environmental tobacco smoke (ETS)
  has been recognized as a cause of human cancer by
  the US Surgeon General,1 the National Institute
  for Occupational Safety and Health,2 the US
  Environmental Protection Agency,3 the California
  Environmental Protection Agency,4 the National
  Health and Medical Research Council of
  Australia,5 the Great Britain Department of
  Health,6 and most recently, the International
  Agency for Research on Cancer.7 Evidence for this
  association has come primarily from studies of
  nonsmokers who are married to a smoker, and
  meta-analyses of these studies have demonstrated
  strong and consistent evidence for an
  association.3,8,9
• Demonstrating an association between workplace
  ETS exposure and lung cancer risk has been more
  difficult. Early meta-analyses failed to
  demonstrate an association between workplace ETS
  exposure and lung cancer risk among
  nonsmokers,1014 but a statistically significant
  association has been reported in the 3 most
  recently published meta-analyses.1517 We sought
  to extend the previous meta-analyses by including
  additional studies and by conducting analyses
  stratified by level of exposure, which was not
  performed in the previous meta-analyses.

95
Does the study approach address the limitations
of previous studies in a logical/compelling way?

• Exposure to environmental tobacco smoke (ETS)
  has been recognized as a cause of human cancer by
  the US Surgeon General,1 the National Institute
  for Occupational Safety and Health,2 the US
  Environmental Protection Agency,3 the California
  Environmental Protection Agency,4 the National
  Health and Medical Research Council of
  Australia,5 the Great Britain Department of
  Health,6 and most recently, the International
  Agency for Research on Cancer.7 Evidence for this
  association has come primarily from studies of
  nonsmokers who are married to a smoker, and
  meta-analyses of these studies have demonstrated
  strong and consistent evidence for an
  association.3,8,9
• Demonstrating an association between workplace
  ETS exposure and lung cancer risk has been more
  difficult. Early meta-analyses failed to
  demonstrate an association between workplace ETS
  exposure and lung cancer risk among
  nonsmokers,1014 but a statistically significant
  association has been reported in the 3 most
  recently published meta-analyses.1517 We sought
  to extend the previous meta-analyses by including
  additional studies and by conducting analyses
  stratified by level of exposure, which was not
  performed in the previous meta-analyses.

96
Does the study approach address the limitations
of previous studies in a logical/compelling way?

• Exposure to environmental tobacco smoke (ETS)
  has been recognized as a cause of human cancer by
  the US Surgeon General,1 the National Institute
  for Occupational Safety and Health,2 the US
  Environmental Protection Agency,3 the California
  Environmental Protection Agency,4 the National
  Health and Medical Research Council of
  Australia,5 the Great Britain Department of
  Health,6 and most recently, the International
  Agency for Research on Cancer.7 Evidence for this
  association has come primarily from studies of
  nonsmokers who are married to a smoker, and
  meta-analyses of these studies have demonstrated
  strong and consistent evidence for an
  association.3,8,9
• Demonstrating an association between workplace
  ETS exposure and lung cancer risk has been more
  difficult. Early meta-analyses failed to
  demonstrate an association between workplace ETS
  exposure and lung cancer risk among
  nonsmokers,1014 but a statistically significant
  association has been reported in the 3 most
  recently published meta-analyses.1517 We sought
  to extend the previous meta-analyses by including
  additional studies and by conducting analyses
  stratified by level of exposure, which was not
  performed in the previous meta-analyses.

97
Why will filling this gap be useful?

• Exposure to environmental tobacco smoke (ETS)
  has been recognized as a cause of human cancer by
  the US Surgeon General,1 the National Institute
  for Occupational Safety and Health,2 the US
  Environmental Protection Agency,3 the California
  Environmental Protection Agency,4 the National
  Health and Medical Research Council of
  Australia,5 the Great Britain Department of
  Health,6 and most recently, the International
  Agency for Research on Cancer.7 Evidence for this
  association has come primarily from studies of
  nonsmokers who are married to a smoker, and
  meta-analyses of these studies have demonstrated
  strong and consistent evidence for an
  association.3,8,9
• Demonstrating an association between workplace
  ETS exposure and lung cancer risk has been more
  difficult. Early meta-analyses failed to
  demonstrate an association between workplace ETS
  exposure and lung cancer risk among
  nonsmokers,1014 but a statistically significant
  association has been reported in the 3 most
  recently published meta-analyses.1517 We sought
  to extend the previous meta-analyses by including
  additional studies and by conducting analyses
  stratified by level of exposure, which was not
  performed in the previous meta-analyses.

98
Reducing the Illegal Sale of Cigarettes to Minors

• Group Exercise
• Read Abstract and Introduction
• What is the gap?

99

• What gap are the authors trying to fill?

100
(No Transcript)
101
Gap Although most states, including California,
have laws regulating the access of minors to
tobacco, they are rarely enforced. By all
accounts, tobacco is readily accessible to
minors. GB In field trials, minors have
successfully purchased tobacco from stores and
vending machines 70 to 100 of the time.
National survey data indicate that 57 of high
school seniors who report daily smoking began by
the age of 14 years. GU Stopping the sale
of tobacco to minors through enforcement of
existing laws is, therefore, a critical step in
any effort to prevent tobacco use. What
about the limitations of previous studies (GL)?
102
Gap Although most states, including California,
have laws regulating the access of minors to
tobacco, they are rarely enforced. By all
accounts, tobacco is readily accessible to
minors. GB In field trials, minors have
successfully purchased tobacco from stores and
vending machines 70 to 100 of the time.
National survey data indicate that 57 of high
school seniors who report daily smoking began by
the age of 14 years. GU Stopping the sale
of tobacco to minors through enforcement of
existing laws is, therefore, a critical step in
any effort to prevent tobacco use. AB We
are aware of only two published studies that
measured the effect of efforts to enforce
existing laws against the sale of cigarettes to
minors. In Decatur, IL, a voluntary merchant
education program reduced the ability of minors
to purchase cigarettes from stores by 18. In
Avon, England, a voluntary merchant education
program combined with a media campaign reduced
the ability of minors to purchase tobacco from
91 of tobacconists to 44. Approach The
purpose of the current study was to determine the
effectiveness of a program to reduce illegal
sales of cigarettes to minors that includes
AN(L) three components voluntary merchant
education, a media campaign, and grassroots
community organization.
103

• Original 500 words
• Revision 200 words

104
(No Transcript)
105
(No Transcript)
106
(No Transcript)
107
(No Transcript)
108
(No Transcript)
109
(No Transcript)
110
(No Transcript)
111
(No Transcript)
112
(No Transcript)
113
Passive voice exercise

• On the basis of evidence from clinical trials,
  the Advisory Committee on Immunization Practices
  (ACIP) recommended in February 2000 that a 4-dose
  PCV7 regimen should be given to all children aged
  lt2 years.
• Try to convert this to active voice without
  changing the meaning.

114
Passive voice exercise

• On the basis of evidence from clinical trials,
  the Advisory Committee on Immunization Practices
  (ACIP) recommended in February 2000 that a 4-dose
  PCV7 regimen should be given to all children aged
  lt2 years.

115
Passive voice exercise

• On the basis of evidence from clinical trials,
  the Advisory Committee on Immunization Practices
  (ACIP) recommended in February 2000 that a 4-dose
  PCV7 regimen should be given to all children aged
  lt2 years.
• On the basis of evidence from clinical trials,
  the Advisory Committee on Immunization Practices
  (ACIP) recommended in February 2000 that all
  children aged lt2 years should receive a 4-dose
  PCV7 regimen should be given to.

116
Passive voice exercise

• On the basis of evidence from clinical trials,
  the Advisory Committee on Immunization Practices
  (ACIP) recommended in February 2000 that a 4-dose
  PCV7 regimen should be given to all children aged
  lt2 years.
• On the basis of evidence from clinical trials,
  the Advisory Committee on Immunization Practices
  (ACIP) recommended in February 2000 that all
  children aged lt2 years should receive a 4-dose
  PCV7 regimen should be given to.
• Can you find an extraneous word in this revision?

117

• On the basis of evidence from clinical trials,
  the Advisory Committee on Immunization Practices
  (ACIP) recommended in February 2000 that a 4-dose
  PCV7 regimen should be given to all children aged
  lt2 years.
• On the basis of evidence from clinical trials,
  the Advisory Committee on Immunization Practices
  (ACIP) recommended in February 2000 that all
  children aged lt2 years should receive a 4-dose
  PCV7 regimen should be given to.
• On the basis of evidence from clinical trials,
  the Advisory Committee on Immunization Practices
  (ACIP) recommended in February 2000 that all
  children aged lt2 years should receive a 4-dose
  PCV7 regimen should be given to.

118
Methods Excerpt from Neonatal Intensive Care Nurse Stressors An American Study All NICU nurses in an upper Midwestern state were invited to participate in this study (N167) 59 nurses responded, yielding a 35 response rate. The study sample consisted of 58 females 46 (78) were married and 12 (20) were single. The mean age was 34.74 years (SD5.97). Thirty-four (58) nurses were baccalaureate prepared, compared to 18 (30) who were diploma nurses and 4 (7) who held associate degrees. The mean for years practiced in NICU was 9.2 (SD5.39). This compares to a mean of 12.02 (SD5.76) for years practiced as a RN.
119
Methods Excerpt from Neonatal Intensive Care Nurse Stressors An American Study All NICU nurses in an upper Midwestern state were invited to participate in this study (N167) 59 nurses responded, yielding a 35 response rate.
120

• All NICU nurses in an upper Midwestern state
  were invited to participate in this study
  (N167).
• Participants in this study were randomly chosen
  from health care units located in the upper
  Midwest of the United States.

121

• All NICU nurses in an upper Midwestern state were
  invited to participate in this study (N167).
• Participants in this study were randomly chosen
  from health care units located in the upper
  Midwest of the United States.

122

• All NICU nurses in an upper Midwestern state were
  invited to participate in this study (N167).
• Participants in this study were randomly chosen
  from health care units located in the upper
  Midwest of the United States.

123
Method Data sources. Behavioral Risk Factor
Surveillance System (BRFSS). Cross-sectional
general population survey data from 1987 to 1990,
collected by the BRFSS, were analyzed. The number
of participating states (including the District
of Columbia) increased from 33 in 1987 to 45 in
1990. The primary objective of this system is to
obtain state specific prevalences of behavioral
risks in adult populations that are associated
with the leading causes of death and
morbidity. Using random-digit dialing, each
state selects a random sample of its
noninstitutionalized adult population (age 18 or
older) who have a telephone. For instance, in
1990, thirty-seven states used a multistage
cluster-sampling design based on the Waksberg
method. A trained interviewer reaches a
representative household, randomly selects one
adult 18 years or older and administers a
standard questionnaire (See Marks et al, 1985 or
Remington et al, 1988, for more detailed
description of method). To meet individual needs,
states may choose to use different sampling
methods, such as simple random or stratified
sample designs. The interviewing instrument
consisted of three parts 1) a core of questions
asked by all states, 2) standardized modules of
questions that are developed by CDC and added at
each states discretion, and 3) state questions
that are developed at the discretion of each
participating state to meet its specific needs.
When possible, questions have been adopted from
national surveys, such as the National Health and
Nutrition Examination Surveys (NHANES) and the
National Health Interview Surveys (NHIS). The
edited data were weighted to the age-, race-, and
sex- specific population counts from the most
current census (or intercensal estimate) in each
state, as well as for the respondents
probability of selection. The total sample
available in 1987 was 50080 in 1988, 56447 in
1989, 66867 and in 1990, 81557.
31
124
Method Data sources. Behavioral Risk Factor
Surveillance System (BRFSS). Cross-sectional
general population survey data from 1987 to 1990,
collected by the BRFSS, were analyzed. The number
of participating states (including the District
of Columbia) increased from 33 in 1987 to 45 in
1990. The primary objective of this system is to
obtain state specific prevalences of behavioral
risks in adult populations that are associated
with the leading causes of death and
morbidity. Using random-digit dialing, each
state selects a random sample of its
noninstitutionalized adult population (age 18 or
older) who have a telephone. For instance, in
1990, thirty-seven states used a multistage
cluster-sampling design based on the Waksberg
method. A trained interviewer reaches a
representative household, randomly selects one
adult 18 years or older and administers a
standard questionnaire (See Marks et al, 1985 or
Remington et al, 1988, for more detailed
description of method). To meet individual needs,
states may choose to use different sampling
methods, such as simple random or stratified
sample designs. The interviewing instrument
consisted of three parts 1) a core of questions
asked by all states, 2) standardized modules of
questions that are developed by CDC and added at
each states discretion, and 3) state questions
that are developed at the discretion of each
participating state to meet its specific needs.
When possible, questions have been adopted from
national surveys, such as the National Health and
Nutrition Examination Surveys (NHANES) and the
National Health Interview Surveys (NHIS). The
edited data were weighted to the age-, race-, and
sex- specific population counts from the most
current census (or intercensal estimate) in each
state, as well as for the respondents
probability of selection. The total sample
available in 1987 was 50080 in 1988, 56447 in
1989, 66867 and in 1990, 81557.
31
125
Analysis variables. Definition of hypertension.
Hypertension case was defined by a yes to the
question Have you ever been told by a doctor,
nurse, or other health professional that you have
high blood pressure? and more than once to
Have you been told on one or more occasions that
your blood pressure was high, or have you been
told this only once? Definition of employment
status. We defined employment status (employed,
unemployed less than one year, and unemployed
more than one year) by the respondent
self-classifying him/herself as employed for
wages self-employed out of work for less
than one year homemaker student or
retired. Respondents who identified themselves
as a homemaker, student, or retired (not in the
labor force) were excluded from the analysis.
Those who were self-employed or employed for
wages were considered employed. The two
categories of unemployment represented two levels
of exposure. Definition of covariates. Ethnic
group. The three ethnic/racial groups used were
white, African-American or black, and Hispanic.
Two questions on the BRFSS defined these groups
(1) What is your race? Would you say White
black Asian, pacific islander Aleutian Eskimo
or American Indian or Other and (2) Are you
of Hispanic origin, such as Mexican American
Latin American Puerto Rican or Cuban? yes
no dont know/not sure refused. White was
defined as a response of white to (1) and a
no, dont know/not sure, or refused to (2)
African American by Black to (1) and no,
dont know/not sure or refused to (2) and
Hispanic by White or Black to (1) and yes
to (2). Age group. Each respondent was assigned
to one of five age groups (18 to 30 years old
30-39 years 40-49 years 50-59 years and 60 or
more years) by using the answer to the question
How old were you on your last birthday?
32
126
Alcohol consumption. A measure of alcohol
consumption was derived using the following BRFSS
questions that asked about number of days per
month or week alcohol was consumed and number of
drinks on each occasion. In 1987 and 1988, the
questions asked specifically about beer, wine,
and spirits, while 1989 and 1990 combined these
categories of alcoholic beverages into single
items. The procedure for computing consumption of
ounces of ethanol per week was structured on the
basis of these two approaches. For all years the
first question established whether the respondent
consumed alcohol regularly Have you had any
beer, wine, wine coolers, cocktails, or liquors
during the past month, that is since ____? yes,
no. For 1987 and 1988, two questions for each
category were asked During the past month, how
many days per week or per month did you drink __
? and On the days when you drank __, about
how many beers did you drink on the average?.
The questions for 1989 and 1990 were asked
similarly, except that the second question
consisted of A drink is one can or bottle of
beer, one glass of wine, one can or bottle of
wine cooler, one cocktail, or one shot of liquor.
On days when you drank, about how many drinks did
you drink on the average?. On the basis of
responses to these questions, an estimate of
ounces of ethanol was computed. We assumed that
.16 oz of ethanol were consumed in one drink,
using as a proportion of all drinks consumed
being 45 beer (for 1989 and 1990) (.04oz of
ethanol per drink), 35 wine (.15 oz), and 20
spirits (.45 oz). Six consumption categories were
used after Laforge et al (1990) None, gt0-1
ounces per week, gt2-4, gt4-10 oz, gt10 oz.
33
127
Alcohol consumption. A measure of alcohol
consumption was derived using the following BRFSS
questions that asked about number of days per
month or week alcohol was consumed and number of
drinks on each occasion. In 1987 and 1988, the
questions asked specifically about beer, wine,
and spirits, while 1989 and 1990 combined these
categories of alcoholic beverages into single
items. The procedure for computing consumption of
ounces of ethanol per week was structured on the
basis of these two approaches. For all years the
first question established whether the respondent
consumed alcohol regularly Have you had any
beer, wine, wine coolers, cocktails, or liquors
during the past month, that is since ____? yes,
no. For 1987 and 1988, two questions for each
category were asked During the past month, how
many days per week or per month did you drink __
? and On the days when you drank __, about
how many beers did you drink on the average?.
The questions for 1989 and 1990 were asked
similarly, except that the second question
consisted of A drink is one can or bottle of
beer, one glass of wine, one can or bottle of
wine cooler, one cocktail, or one shot of liquor.
On days when you drank, about how many drinks did
you drink on the average?. On the basis of
responses to these questions, an estimate of
ounces of ethanol was computed. We assumed that
.16 oz of ethanol were consumed in one drink,
using as a proportion of all drinks consumed
being 45 beer (for 1989 and 1990) (.04oz of
ethanol per drink), 35 wine (.15 oz), and 20
spirits (.45 oz). Six consumption categories were
used after Laforge et al (1990) None, gt0-1
ounces per week, gt2-4, gt4-10 oz, gt10 oz.
!
300 words
33
128
(No Transcript)
129
(No Transcript)
130
Value of flow diagrams in reports of randomized
controlled trials JAMA 2001285 19961999
131
(No Transcript)
132
(No Transcript)
133
The METHODS

• Purpose to describe how you collected,
  organized and analyzed the data
• Describe what you did, not what you found
  (Results)
• Respect chronology
• Describe original methods in detail otherwise
  give references.
• (length varies depending on originality of
  methods used)

134
The METHODS

• Purpose to describe how you collected,
  organized and analyzed the data (that are
  relevant to the purpose of the study)
• Clearly present and define all analysis
  variables.
• Organize into logical subsections that illustrate
  the steps you took to collect, organize and
  analyze the data.
• Describe what you did, not what you found
  (Results)
• Respect chronology
• Describe original methods in detail otherwise
  give references.
• (length varies depending on originality of
  methods used)

135
Analysis variables. Definition of hypertension.
Hypertension case was defined by a yes to the
question Have you ever been told by a doctor,
nurse, or other health professional that you have
high blood pressure? and more than once to
Have you been told on one or more occasions that
your blood pressure was high, or have you been
told this only once? Definition of employment
status. We defined employment status (employed,
unemployed less than one year, and unemployed
more than one year) by the respondent
self-classifying him/herself as employed for
wages self-employed out of work for less
than one year homemaker student or
retired. Respondents who identified themselves
as a homemaker, student, or retired (not in the
labor force) were excluded from the analysis.
Those who were self-employed or employed for
wages were considered employed. The two
categories of unemployment represented two levels
of exposure. Definition of covariates. Ethnic
group. The three ethnic/racial groups used were
white, African-American or black, and Hispanic.
Two questions on the BRFSS defined these groups
(1) What is your race? Would you say White
black Asian, pacific islander Aleutian Eskimo
or American Indian or Other and (2) Are you
of Hispanic origin, such as Mexican American
Latin American Puerto Rican or Cuban? yes
no dont know/not sure refused. White was
defined as a response of white to (1) and a
no, dont know/not sure, or refused to (2)
African American by Black to (1) and no,
dont know/not sure or refused to (2) and
Hispanic by White or Black to (1) and yes
to (2). Age group. Each respondent was assigned
to one of five age groups (18 to 30 years old
30-39 years 40-49 years 50-59 years and 60 or
more years) by using the answer to the question
How old were you on your last birthday?
32
136
Reliability of Information on Chronic Disease
Risk Factors Collected in the Missouri Behavioral
Risk Factor Surveillance System
Are the analysis variables clearly defined?
137

• If we read references 16-20, what can we expect
  to learn about the analysis variables?
• how to measure/interpret agreement?
• how to measure/interpret reliability?
• whether reliability and agreement mean the same
  thing?
• how Pearson and Spearman correlation coefficients
  are used to measure agreement?

138
The METHODS

• Purpose to describe how you collected,
  organized and analyzed the data (that are
  relevant to the purpose of the study)

139
The METHODS

• Purpose to describe how you collected,
  organized and analyzed the data (that are
  relevant to the purpose of the study)
• Clearly present and define all analysis
  variables.

140
The METHODS

• Purpose to describe how you collected,
  organized and analyzed the data (that are
  relevant to the purpose of the study)
• Clearly present and define all analysis
  variables.
• Organize into logical subsections that illustrate
  the steps you took to collect, organize and
  analyze the data.

141
Examples of Subsection Headings

• Study population
• How many subjects were eligible (eligibilty
  criteria)
• How many declined to participate (and why)
• How many participated
• How many dropped out
• Sampling methods
• Laboratory methods
• Epidemiologic investigation
• Baseline data collection
• Diagnostic Evaluation
• Intervention
• Follow-up
• Definition of variables
• Exposure variables
• Outcome variables
• Case definition
• St