enduring understandings 7-9 explaining associations and judging causation
Post on 11-Jan-2016
222 Views
Preview:
TRANSCRIPT
Wood County Schools1210 13th Street
Parkersburg, WV 26101June 22-26, 2009
Teach EpidemiologyProfessional Development Workshop
Enduring Understandings 7-9
Explaining associations Explaining associations
and and
judging causationjudging causation
EU7: One possible explanation for EU7: One possible explanation for finding an association is that the finding an association is that the exposure causes the outcome. exposure causes the outcome. Because studies are complicated Because studies are complicated by factors not controlled by the by factors not controlled by the observer, other explanations also observer, other explanations also must be considered, including must be considered, including confounding, chance, and bias.confounding, chance, and bias.
The “Not everything that glitters is The “Not everything that glitters is gold” Principlegold” Principle
EU8: Judgments about whether an EU8: Judgments about whether an exposure causes a disease are exposure causes a disease are developed by examining a body of developed by examining a body of epidemiologic evidence, as well as epidemiologic evidence, as well as evidence from other scientific evidence from other scientific disciplines.disciplines.
EU9: While a given exposure may be EU9: While a given exposure may be necessary to cause an outcome, the necessary to cause an outcome, the presence of a single factor is seldom presence of a single factor is seldom sufficient. Most outcomes are caused sufficient. Most outcomes are caused by a combination of exposures that may by a combination of exposures that may include genetic make-up, behaviors, include genetic make-up, behaviors, social, economic, and cultural factors social, economic, and cultural factors and the environment. and the environment.
The “Just because your friend sleeps in The “Just because your friend sleeps in class and never fails her courses does class and never fails her courses does not mean that sleeping in class does not not mean that sleeping in class does not cause F grades” Principlecause F grades” Principle
Reasons for associations ConfoundingConfounding
E is associated with C and C causes DE is associated with C and C causes D BiasBias
F causes D, but we thought F was an EF causes D, but we thought F was an E Reverse causalityReverse causality
““D” causes “E”D” causes “E” Sampling error (chance)Sampling error (chance) CausationCausation
E1 E1 DD E1 + E2 E1 + E2 D D E1 or E2 E1 or E2 D D E1 + E2 E1 + E2 OROR E3+E4 E3+E4DD
Osteoporosis risk is higher among Osteoporosis risk is higher among women who live alone. women who live alone.
Death rates are low in AK and high Death rates are low in AK and high in FL.in FL.
African American women have African American women have higher infant mortality than others higher infant mortality than others in the US.in the US.
Confounding
Confounding is an alternate Confounding is an alternate explanation for an observed explanation for an observed association of interest.association of interest.Number of
persons in the home
Osteoporosis
Age
Confounding
Confounding is an alternate Confounding is an alternate explanation for an observed explanation for an observed association of interest.association of interest.Exposure Outcome
Confounder
Confounding
YES confounding module YES confounding module example:example:Hypothetical cohort studyHypothetical cohort study20,000 men followed for 10 yrs20,000 men followed for 10 yrsRQ: Are bedsores related to RQ: Are bedsores related to mortality among elderly mortality among elderly patients with hip fractures?patients with hip fractures?
Bedsores and Mortality
D+D+ D-D-
E+E+ 7979 745745 824824
E-E- 286286 82908290 85768576
365365 90359035 94009400
RR = (79 / 824) / (286 / 8576) = 2.9
Bedsores and Mortality
Avoid bedsores…Live Avoid bedsores…Live forever!!forever!!
Could there be some Could there be some other explanation for the other explanation for the observed association?observed association?
Bedsores and mortality
If severity of medical problems had If severity of medical problems had been the reason for the association been the reason for the association between bedsores and mortality, between bedsores and mortality, what might the RR be if all study what might the RR be if all study participants had very severe participants had very severe medical problems?medical problems?
What about if the participants all What about if the participants all had problems of very low severity?had problems of very low severity?
Bedsores and Mortality
DiedDied Did not Did not diedie
BedsoresBedsores 55 severe55 severe
24 not24 not51 severe51 severe
694 not694 not824824
No No bedsoresbedsores
5 severe5 severe
281 not281 not5 severe5 severe
8285 not8285 not85768576
365365 90359035 94009400
Bedsores and Mortality (Severe)
DiedDied Did not Did not diedie
BedsoresBedsores 5555 5151 106106
No No bedsoresbedsores
55 55 1010
6060 5656 116116
RR = (55 / 106) / (5 / 10) = 1.0
Bedsores and Mortality (Not severe)
DiedDied Did not Did not diedie
BedsoresBedsores 2424 694694 718718
No No bedsoresbedsores
281281 82858285 85668566
305305 89798979 92849284
RR = (24 / 718) / (281 / 8566) = 1.0
Bedsores and Mortality stratified by Medical SeveritySEVERESEVERE++
DiedDied Didn’t Didn’t diedie
BedsoresBedsores aa bb
No soresNo sores cc dd
RR = RR = 1.01.0
SEVERSEVERE-E-
DiedDied Didn’t Didn’t diedie
BedsoresBedsores aa bb
No soresNo sores cc dd
RR = RR = 1.01.0
SEVERESEVERE++
DiedDied Didn’t Didn’t diedie
BedsoresBedsores aa bb
No soresNo sores cc dd
RR = RR = 2.92.9
SEVERSEVERE-E-
DiedDied Didn’t Didn’t diedie
BedsoresBedsores aa bb
No soresNo sores cc dd
RR = RR = 2.92.9
Bedsores
So….So…. Bedsores are unrelated to mortality Bedsores are unrelated to mortality
among those with severe problems.among those with severe problems. Bedsores are unrelated to mortality Bedsores are unrelated to mortality
among those with problems of less among those with problems of less severity.severity.
…….. the adjusted RR = 1, and the unadjusted the adjusted RR = 1, and the unadjusted
RR = 2.9RR = 2.9
Confounding
Confounding is an alternate Confounding is an alternate explanation for an observed explanation for an observed association of interest.association of interest.Bedsores Death
Severity of medical problems
Reasons for associations ConfoundingConfounding
E is associated with C and C causes DE is associated with C and C causes D BiasBias
F causes D, but we thought F was an EF causes D, but we thought F was an E Reverse causalityReverse causality
““D” causes “E”D” causes “E” Sampling error (chance)Sampling error (chance) CausationCausation
E1 E1 DD E1 + E2 E1 + E2 D D E1 or E2 E1 or E2 D D E1 + E2 E1 + E2 OROR E3+E4 E3+E4DD
Bias
Errors are mistakes that are:Errors are mistakes that are: randomly distributedrandomly distributed not expected to impact the MAnot expected to impact the MA less modifiableless modifiable
Biases are mistakes that are:Biases are mistakes that are: not randomly distributednot randomly distributed may impact the MAmay impact the MA more modifiablemore modifiable
Types of bias
Selection biasSelection bias The process for The process for selecting/keeping selecting/keeping
subjects causes mistakessubjects causes mistakes Information biasInformation bias
The process for collecting The process for collecting informationinformation from the subjects from the subjects causes mistakescauses mistakes
Selection bias Healthy worker effectHealthy worker effect
People who are working are more likely to People who are working are more likely to be healthier than non-workersbe healthier than non-workers
Non-responseNon-response People who participate in a study may be People who participate in a study may be
different from people who do notdifferent from people who do not AttritionAttrition
People who drop out of a study may be less People who drop out of a study may be less different from those who stay in the studydifferent from those who stay in the study
Berkson’sBerkson’s Hospital controls in a case-control studyHospital controls in a case-control study
Information bias
Misclassification, e.g. non-exposed as Misclassification, e.g. non-exposed as exposed or cases as controlsexposed or cases as controls
Recall biasRecall bias Cases are more likely than controls Cases are more likely than controls
to recall past exposuresto recall past exposures Interviewer biasInterviewer bias
Interviewers probe cases more than Interviewers probe cases more than controls (exposed more than controls (exposed more than unexposed)unexposed)
Birth defects and diet
In a study of birth defects, mothers In a study of birth defects, mothers of children with and without of children with and without infantile cataracts are asked about infantile cataracts are asked about dietary habits during pregnancy.dietary habits during pregnancy.
Pesticides and cancer mortality In a study of the relationship In a study of the relationship
between home pesticide use and between home pesticide use and cancer mortality, controls are cancer mortality, controls are asked about pesticide use and asked about pesticide use and family members are asked about family members are asked about their loved ones’ usage patterns.their loved ones’ usage patterns.
Induced abortion & breast CA Positive association found in 5 Positive association found in 5
studiesstudies No association found in 6 studiesNo association found in 6 studies Negative association found in 1 Negative association found in 1
studystudy
Minimize bias
Can only be done in the planning and Can only be done in the planning and implementation phaseimplementation phase
Standardized processes for data Standardized processes for data collectioncollection
MaskingMasking Clear, comprehensive case definitionsClear, comprehensive case definitions Incentives for participation/retentionIncentives for participation/retention
Reasons for associations ConfoundingConfounding
E is associated with C and C causes DE is associated with C and C causes D BiasBias
F causes D, but we thought F was an EF causes D, but we thought F was an E Reverse causalityReverse causality
““D” causes “E”D” causes “E” Sampling error (chance)Sampling error (chance) CausationCausation
E1 E1 DD E1 + E2 E1 + E2 D D E1 or E2 E1 or E2 D D E1 + E2 E1 + E2 OROR E3+E4 E3+E4DD
Reverse causality
Suspected disease actually precedes Suspected disease actually precedes suspected causesuspected cause
Pre-clinical disease Pre-clinical disease Exposure Exposure Disease Disease For example: Memory deficits For example: Memory deficits
Reading cessation Reading cessation Alzheimer’s Alzheimer’s Cross-sectional studyCross-sectional study
For example: Sexual For example: Sexual activity/Marijuanaactivity/Marijuana
Minimize effect of reverse causality Done in the planning and Done in the planning and
implementation phase of a studyimplementation phase of a study Pick study designs in which Pick study designs in which
exposure is measured before exposure is measured before disease onsetdisease onset
Assess disease status with as Assess disease status with as much accuracy as possiblemuch accuracy as possible
Reasons for associations ConfoundingConfounding
E is associated with C and C causes DE is associated with C and C causes D BiasBias
F causes D, but we thought F was an EF causes D, but we thought F was an E Reverse causalityReverse causality
““D” causes “E”D” causes “E” Sampling error (chance)Sampling error (chance) CausationCausation
E1 E1 DD E1 + E2 E1 + E2 D D E1 or E2 E1 or E2 D D E1 + E2 E1 + E2 OROR E3+E4 E3+E4DD
Sampling error/chance
E and D are associated in a E and D are associated in a sample, but not in the population sample, but not in the population from which the sample was drawn.from which the sample was drawn.
RR in the population
D+D+ D-D-
E+E+ 5050 5050 100100
E-E- 5050 5050 100100
100100 100100 200200
RR in sample1
D+D+ D-D-
E+E+ 2525 2525 5050
E-E- 2525 2525 5050
5050 5050 100100
RR in sample2
D+D+ D-D-
E+E+ 2020 3030 5050
E-E- 3030 2020 5050
5050 5050 100100
RR in sample3
D+D+ D-D-
E+E+ 3030 2020 5050
E-E- 1515 3535 5050
4545 5555 100100
Reasons for associations ConfoundingConfounding
E is associated with C and C causes DE is associated with C and C causes D BiasBias
F causes D, but we thought F was an EF causes D, but we thought F was an E Reverse causalityReverse causality
““D” causes “E”D” causes “E” Sampling error (chance)Sampling error (chance) CausationCausation
E1 E1 DD E1 + E2 E1 + E2 D D E1 or E2 E1 or E2 D D E1 + E2 E1 + E2 OROR E3+E4 E3+E4DD
Causal pathways
Necessary, sufficient—rare, if at allNecessary, sufficient—rare, if at all Not necessary, sufficient—also rareNot necessary, sufficient—also rare Necessary, not sufficient—TBNecessary, not sufficient—TB Not necessary, not sufficient--Most Not necessary, not sufficient--Most
causes fall into this category--heart causes fall into this category--heart disease, obesitydisease, obesity
Reasons for associations ConfoundingConfounding
E is associated with C and C causes DE is associated with C and C causes D BiasBias
F causes D, but we thought F was an EF causes D, but we thought F was an E Reverse causalityReverse causality
““D” causes “E”D” causes “E” Sampling error (chance)Sampling error (chance) CausationCausation
E1 E1 DD E1 + E2 E1 + E2 D D E1 or E2 E1 or E2 D D E1 + E2 E1 + E2 OROR E3+E4 E3+E4DD
The process of assessing causality Observe patternsObserve patterns Generate hypothesisGenerate hypothesis Design study to test hypothesisDesign study to test hypothesis Conduct studyConduct study Interpret the results…the big question is did Interpret the results…the big question is did
the exposure cause the disease?the exposure cause the disease? Are there alternate non-causal Are there alternate non-causal
explanations for the results we explanations for the results we found?found?
If not, then is this the whole story?If not, then is this the whole story?
So, what should we do?
Goal is to understand causalityGoal is to understand causality Use guidelines to help us make Use guidelines to help us make
sense of the evidencesense of the evidence
Key Guidelines
Temporality: a necessary conditionTemporality: a necessary condition ConsistencyConsistency Dose-responseDose-response Consideration of alternate Consideration of alternate
explanationsexplanations CoherenceCoherence
Enduring Understandings
7, 8, and 97, 8, and 9
EU7: One possible explanation for EU7: One possible explanation for finding an association is that the finding an association is that the exposure causes the outcome. exposure causes the outcome. Because studies are complicated Because studies are complicated by factors not controlled by the by factors not controlled by the observer, other explanations also observer, other explanations also must be considered, including must be considered, including confounding, chance, and bias.confounding, chance, and bias.
The “Not everything that glitters is The “Not everything that glitters is gold” Principlegold” Principle
EU8: Judgments about whether an EU8: Judgments about whether an exposure causes a disease are exposure causes a disease are developed by examining a body of developed by examining a body of epidemiologic evidence, as well as epidemiologic evidence, as well as evidence from other scientific evidence from other scientific disciplines.disciplines.
EU9: While a given exposure may be EU9: While a given exposure may be necessary to cause an outcome, the necessary to cause an outcome, the presence of a single factor is seldom presence of a single factor is seldom sufficient. Most outcomes are caused sufficient. Most outcomes are caused by a combination of exposures that may by a combination of exposures that may include genetic make-up, behaviors, include genetic make-up, behaviors, social, economic, and cultural factors social, economic, and cultural factors and the environment. and the environment.
The “Just because your friend sleeps in The “Just because your friend sleeps in class and never fails her courses does class and never fails her courses does not mean that sleeping in class does not not mean that sleeping in class does not cause F grades” Principlecause F grades” Principle
49
1. Cause
2. Confounding
3. Reverse Time Order
4. Chance
5. Bias
Possible Explanations for Finding an Association
50
Cause
A factor that produces a change in another factor.
William A. Oleckno, Essential Epidemiology: Principles and Applications, Waveland Press, 2002.
Possible Explanations for Finding an Association
51
Sample of 100
52
Sample of 100, 25 are Sick
53
Diagram
2x2 Table
DZ DZ
X
X
a bc d
Types of Causal Relationships
54
DZ DZ
X
X
a bc d
Diagram
2x2 Table
Types of Causal Relationships
55
Handout
56
X
X
X
X
X
X
X
X
X X
X
XXX
XX
X
X
X
X
X
X
X
X
X
X DZ
DZ DZ
X
X
a bc d
X
Diagram
2x2 Table
Necessary and Sufficient
57
DZ DZ
X
X
a bc d
X DZX X+ +
X
X
X
X
X
X
X
X
X
X X
X
XXX
XX
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
XX
Diagram
2x2 Table
Necessary but Not Sufficient
58
X
X
X
X
X
X
X
X X
X
XX
X
X
X
X
DZ DZ
X
X
a bc d
X
X DZ
X
X
Diagram
2x2 Table
Not Necessary but Sufficient
59
DZ DZ
X
X
a bc d
X
X
X
X
X
X
X
X X
XXX
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
XX
X
X
X
DZX X+ +
X X+ +
X X+ +
Not Necessary and Not Sufficient
Diagram
2x2 Table
60
a b
c d
Heart Attack
NoHeart Attack
Lack of Fitness
No Lack of Fitness
Lack of fitness and physical activity causes heart attacks.
61
a b
c d
Lead Poisoning
NoLead
Poisoning
Lack of Supervision
No Lack of
Supervision
Lack of supervision of small children causes lead poisoning.
62
Is the association causal?
63
Suicide Higher in Areas with Guns
Family Meals Are Good for Mental Health
Lack of High School Diploma Tied to US Death
Rate
Study Links
Spanking to
Aggression
Study Concludes: Movies Influence
Youth Smoking
Study Links Iron
Deficiency to Math
Scores
Kids Who Watch R-Rated Movies More Likely to Drink, Smoke
Pollution Linked with Birth Defects in US Study
Ties, Links, Relationships, and Associations
1. Cause
2. Confounding
3. Reverse Time Order
4. Chance
5. Bias
Snacks Key to Kids’ TV- Linked Obesity: China
Study
Depressed Teens More
Likely to Smoke
64
65
1. Cause
2. Confounding
3. Reverse Time Order
4. Chance
5. Bias
Possible Explanations for Finding an Association
66
All the people in a particular group.
Population
Possible Explanations for Finding an Association
67
A selection of people from a population.
Sample
Possible Explanations for Finding an Association
68
Inference
Process of predicting from what is observed in a sample to what is not observed in a population.
To generalize back to the source population.
Possible Explanations for Finding an Association
69
Sample
Population
Process of predicting from what is observed
to what is not observed.
Observed
Not Observed
Inference
70
Deck of
100 cards
Population
71
a
25 cards
b
25 cards
c
25 cards
25 cards
d
Population
72
=
Population
a
25 cards
b c d
25 cards25 cards25 cards
=a b
c d
Odd #
Even #
No Marijuana
No Marijuana
Population
Total
73
=
Population
a
25 cards
b c d
25 cards25 cards25 cards
= 2525
25 25
50
50
Total
Odd #
Even #
No Marijuana
No Marijuana
Population
74
=
Population
=M&M’s
No M&M’s
FluNo
Flu
2525
25 25
50
50
Total
=
2525
25 25
50
50
Total
a
25 cards
b c d
25 cards25 cards25 cards
Odd #
Even #
No Marijuana
No Marijuana
Population
75
=
Population
=
2525
25 25
50
50
Total
a
25 cards
b c d
25 cards25 cards25 cards
Risk
25 / 50 or 50%
25 / 50 or 50%
Odd #
Even #
No Marijuana
No Marijuana
Population
76
=
Population
a
25 cards
b c d
25 cards25 cards25 cards
=
2525
25 25
50
50
Total Risk Relative Risk
25 / 50 or 50 %
25 / 50 or 50 %50 % / 50% = = 1
50 %
50 %
____Odd #
Even #
No Marijuana
No Marijuana
Population
77
25 cards
25 cards
25 cards
25 cards
Population
78
To occur accidentally.
To occur without design.
Chance
A coincidence.
Possible Explanations for Finding an Association
79
Chance
80
Chance
81
Population
Sample
b
Sample
of
20 cards25 cards25 cards25 cards25 cards
Sample
82
Population
Sample
b
Sample
of
20 cards25 cards25 cards25 cards25 cards
10
10
Total
55
5 5Odd #
Even #
No Marijuana
No Marijuana
Sample
83
Population
Sample
b
Sample
of
20 cards25 cards25 cards25 cards25 cards
10
10
Total
55
5 5
Risk
5 / 10 or 50 %
5 / 10 or 50 %Odd #
Even #
No Marijuana
No Marijuana
Sample
84
Population
Sample
b
Sample
of
20 cards25 cards25 cards25 cards25 cards
10
10
Total
55
5 5
Risk
5 / 10 or 50 %
5 / 10 or 50 %Odd #
Even #
No Marijuana
No Marijuana
Sample
Relative Risk
50 % / 50% = = 150 %
50 %
____
85
b
Sample
of
20 cards
TotalRisk
5 / 10 = 50 %
5 / 10 = 50 %
50 1
Relative Risk
By Chance CDC
% ___
%
=Odd #
Even #
No Marijuana
No Marijuana
Sample
86
10
10
Total
55
5 5
Risk
5 / 10 or 50 %
5 / 10 or 50 %
Relative Risk
How many students picked a sample with 5 people in each cell?
= 150 %
50 %
____
Odd #
Even #
No Marijuana
No Marijuana
Chance
By Chance
87
Suicide Higher in Areas with Guns
Family Meals Are Good for Mental Health
Lack of High School Diploma Tied to US Death
Rate
Study Links
Spanking to
Aggression
Study Concludes: Movies Influence
Youth Smoking
Study Links Iron
Deficiency to Math
Scores
Kids Who Watch R-Rated Movies More Likely to Drink, Smoke
1. Cause
2. Confounding
3. Reverse Time Order
4. Chance
5. Bias
Snacks Key to Kids’ TV- Linked Obesity: China
Study
Depressed Teens More
Likely to Smoke
Association is not necessarily causation.
Ties, Links, Relationships, and Associations
88
An Association: TV and Aggressive ActsAn Association: TV and Aggressive Acts
Worksheet
“… the study of the distribution and determinants of health-related states or events …”
A Study Finds More Links Between TV and Violence
March 29, 2002
By GINA KOLATA
The New York Times ON THE WEB
Study Designs
Experimental Studies
Observational Studies
Randomized Controlled Trials
Other Experimental Studies
Cohort Studies
Case-Control Studies
Cross-Sectional Studies
Ecologic Studies
Cohort Studies
• A study in which a group of people is followed over time
• The group is made up of people who have the exposure of interest and people who do not have the exposure of interest
• Exposed and unexposed people are followed over time to determine whether they experience the outcome
Cohort Study
When epidemiologists ask a question, it is often of the form:
Does ______________ cause ______________?
Exposure - Outcome
(exposure) (outcome)
Do diesel exhaust fumes from school buses cause asthma?
Does eating chocolate cause acne?
Are males at higher risk of automobile accidents?
Does immunization with the measles vaccine prevent measles?
Does acupuncture result in pain relief?
Exposure - Outcome
For example:
When epidemiologists ask a question, it is often of the form:
Does ______________ cause ______________?(exposure) (outcome)
A Study Finds More Links Between TV and
Violence
March 29, 2002
By GINA KOLATA
Cohort Study Flow Diagram
A designated group of persons who are followed or traced over a period of time
Exposed
Not Exposed
Time
-
Cohort
Outcome
No Outcome
Outcome
No Outcome
By age 22
A Study Finds More Links Between TV and
Violence
March 29, 2002
By GINA KOLATA
Cohort Study Flow Diagram
A designated group of persons who are followed or traced over a period of time
Watching TV for > 1
hrs per day
Watching TV for < 1 hr per day
At age 14
-
Adolescents & Young Adults
Aggressive Acts
No Aggressive
Acts
Aggressive Acts
No Aggressive
Acts
Watched TV >
1 hour per day
At age 14
154 reported
aggressive acts
465 did not report
aggressive acts
Express it in Numbers
By age 22
Express it in Numbers
Exposed
Outcome TotalNo
Outcome
Watched TV >
1 hour per day
At age 14
154 reported
aggressive acts
465 did not report
aggressive acts
By age 22
Exposed
Outcome TotalNo
Outcome
At age 14 By age 22
Express it in Numbers
Watched TV
> 1 hour
per day
Aggressive Acts
No
Aggressive Acts
154 465 619
Total
Watched TV >
1 hour per day
154 reported
aggressive acts
465 did not report
aggressive acts
Exposed
Outcome TotalNo
Outcome
Risk
Aggressive Acts
No
Aggressive Acts
154 465 619
Total
154(154 + 465)
=154
619 24.9%=
Watched TV
> 1 hour
per day
An unproven idea, based on observation or reasoning, that can be
proven or disproven through investigation
An educated guess
Hypothesis
Watching TV causes aggressive acts.
Exposed
Outcome TotalNo
OutcomeAggressiv
e Acts
No
Aggressive Acts
154 465 619
Total
Does watching TV cause aggressive acts?
Risk
24.9%Watched TV
> 1 hour
per day
154(154 + 465)
= 24.9%154
619 =
By 22 years
Watching TV for > 1
hrs per day
Watching TV for < 1 hr per day
At 14 years
-
Aggressive Acts
No Aggressive
Acts
Aggressive Acts
No Aggressive
Acts
24.9% risk of
committing an
aggressive act
? risk of
committing an
aggressive act
Does watching TV cause aggressive acts?
Adolescents & Young Adults
By 22 years
Watching TV for > 1
hrs per day
Watching TV for < 1 hr per day
At 14 years
-
Aggressive Acts
No Aggressive
Acts
Aggressive Acts
No Aggressive
Acts
24.9% risk of
committing an
aggressive act
? risk of
committing an
aggressive actComparison
Group
Does watching TV cause aggressive acts?
Adolescents & Young Adults
Exposed
Outcome TotalNo
OutcomeAggressiv
e Acts
No
Aggressive Acts
154 465 619
Total
Comparison Group
Risk
24.9%Watched TV > 1 hour per dayWatched TV < 1 hour per day
5 reported aggressive acts
83 did not report aggressive
acts
Watched TV < 1 hour
per day
At age 14 By age 22
Exposed
Outcome TotalAggressiv
e Acts
No
Aggressive Acts
154 465 619
Total
Comparison Group
Risk
24.9%Watched TV > 1 hour per dayWatched TV < 1 hour per day
5 reported aggressive acts
83 did not report aggressive
acts
Watched TV < 1 hour
per day
At age 14 By age 22
5 83 88 5.7%
Exposed
Outcome TotalNo
OutcomeAggressiv
e Acts
No
Aggressive Acts
154 465 619
Total Risk
24.9%Watched TV > 1 hour per day
Watched TV < 1 hour per day
5 83 88 5.7%Exp
osu
re
Outcome
Contingency Table
Exposed
Outcome TotalNo
OutcomeAggressiv
e Acts
No
Aggressive Acts
154 465 619
Total Risk
24.9%Watched TV > 1 hour per day
Watched TV < 1 hour per day
5 83 88 5.7%
Does watching TV cause aggressive acts?
Exposed
Outcome TotalNo
OutcomeAggressiv
e Acts
No
Aggressive Acts
154 465 619
Total Risk
24.9%Watched TV > 1 hour per day
Watched TV < 1 hour per day
5 83 88 5.7%
Compared to those who watched TV for < 1 hour per day, those who watched TV for > 1 hours per day were ____
times as likely to commit aggressive acts.
4.4
Does watching TV cause aggressive acts?
Times as
Likely
A way of quantifying the relationship between two risks
Tells us the number of times one risk is larger or smaller than another
Relative Risk
Cartoon from Larry Gotnick’s The Cartoon Guide to Statistics, HarperPerennial, 1993
“… the control of health problems”
What should be done?
Exposed
Outcome TotalNo
OutcomeAggressiv
e Acts
No
Aggressive Acts
154 465 619
Total Risk
24.9%Watched TV > 1 hour per day
Watched TV < 1 hour per day
5 83 88 5.7%
4.4
Relative Risk
When things turn up together
Association
Pretzels Auto Accidents
Confounding
Another Exposure
Association Cause
When an observed association between an exposure and an outcome is distorted because the exposure of interest is associated with
some other exposure that causes the outcome
Drinking Alcoholic
Beverages
Association of Interest
• Confounding is the distortion of an exposure-outcome association brought about by the association of another factor with both outcome and exposure.
• A confounder confuses our conclusions about the relationship between an exposure and an outcome.
Confounding
Pretzels Auto Accidents
Another Exposure
Association Cause
“… the control of health problems”
X
Drinking Alcoholic
BeveragesX
Association of Interest
Association
When things turn up together
Aggressive Acts
No
Aggressive Acts
Total
Watched TV < 1 hour per day
Watched TV > 1 hour per day
Relative Risk
619154 465
Total Risk
24.9%
5 83 88 5.7%
4.4
Confounding
Association Cause
?
Watching TV
Aggressive Acts
Association of Interest
When an observed association between an exposure and an outcome is distorted because the exposure of interest is associated with
some other exposure that causes the outcome
Watching TV
Aggressive Acts
Confounding
Association Cause
Living in a Violent
Neighborhood
Association of Interest
When an observed association between an exposure and an outcome is distorted because the exposure of interest is associated with
some other exposure that causes the outcome
Watching TV
Aggressive Acts
Confounding
Association Cause
Lack of Adequate
Supervision
Association of Interest
When an observed association between an exposure and an outcome is distorted because the exposure of interest is associated with
some other exposure that causes the outcome
Watching TV
Aggressive Acts
Association Cause
Lack of Adequate
Supervision
X
X
“… the control of health problems”
Association of Interest
When an observed association between an exposure and an outcome is distorted because the exposure of interest is associated with
some other exposure that causes the outcome
Assessment
In a study of the hypothesis that drinking orange juice prevents the flu, 3,000 students at Wright High School, who did not have the flu on December 31, 2000, were followed from January 1 through March 31, 2001. By the end of the study, among the 1000 students who drank orange juice, 123 students had developed the flu. Among the 2000 students who did not drink orange juice, 342 students had developed the flu. Display the above data on a 2x2 table, calculate risks of flu, calculate the relative risk, and explain whether or not the results support the hypothesis that drinking orange juice prevents the flu.
123
124
Guilt or Innocence?Causal or Not Causal?
Does evidence from an aggregate of studies support a cause-effect relationship?
Teach Epidemiology
Explaining Associations and Judging Causation
125
Sir Austin Bradford Hill “The Environment and Disease:
Association or Causation?” Proceedings of the Royal Society of Medicine
January 14, 1965
Teach Epidemiology
Explaining Associations and Judging Causation
126
“In what circumstances can we pass from this observed association
to a verdict of causation?”
Teach Epidemiology
Explaining Associations and Judging Causation
127
“Here then are nine different viewpoints from all of which we should study association
before we cry causation.”
Teach Epidemiology
Explaining Associations and Judging Causation
Does evidence from an aggregate of studies support a cause-effect relationship?
1. What is the strength of the association between the risk factor and the disease?
2. Can a biological gradient be demonstrated?
3. Is the finding consistent? Has it been replicated by others in other places?
4. Have studies established that the risk factor precedes the disease?
5. Is the risk factor associated with one disease or many different diseases?
6. Is the new finding coherent with earlier knowledge about the risk factor and the m disease?
7. Are the implications of the observed findings biological sensible?
8. Is there experimental evidence, in humans or animals, in which the disease has m been produced by controlled administration of the risk factor?
Teach Epidemiology
Explaining Associations and Judging Causation
129
Stress causes ulcers.
Helicobacter pylori causes ulcers.
Teach Epidemiology
Explaining Associations and Judging Causation
130
*
*
*
**
*
*
*
*
Teach Epidemiology
Explaining Associations and Judging Causation
131Teach Epidemiology
Explaining Associations and Judging Causation
132
In the News
• Assemble into three-person teams• Select an article• Use the article to create a lesson plan to teach
one or more of the Enduring Understandings to a specified class for 30 minutes
• Teach the lesson– Specify the student population and course– Engage us as though we were the students
• Help us to understand what you did to generate the lesson plan
Teach Epidemiology
Article Choices
• Early childhood behavior and substance use• Huffing and suicide• Soft drinks and diabetes• Circumcision and AIDS• Prenatal smoking and attention deficit• ADHD among girls• Traffic and childhood asthma• Breast-feeding and childhood obesity• Depression and sexual risk-taking• Family stress and childhood illness• ADHD medications and mortality
Teach Epidemiology
136
1. Teach epidemiology.
2. As a group, create a 30-minute lesson during which we will develop a deeper understanding of an enduring epidemiological understanding.
3. Focus on the portion of the unit that is assigned. Use that portion of the unit as the starting point for creating your 30-minute lesson.
4. When teaching, assume the foundational epidemiological knowledge from the preceding days of the workshop.
5. Try to get us to uncover the enduring epidemiological understanding. Try to only tell us something when absolutely necessary.
6. End each lesson by placing it in the context of the appropriate enduring epidemiological understanding.
7. Teach epidemiology.
8. Metacognition--After the lesson, reflect on your preparation for and teaching of the lesson.
Teach Epidemiology
Teaching Epidemiology Rules
137
They can then use that ability to think about their own thinking … to grasp how other people might learn. They know what has to come first,
and they can distinguish between foundational concepts and
elaborations or illustrations of those ideas.
They realize where people are likely to face difficulties developing their own comprehension,
and they can use that understanding to simplify
and clarify complex topics for others, tell the right story, or raise a powerfully provocative question.
Ken Bain, What the Best College Teachers Do
Teach Epidemiology
Teaching Epidemiology
Metacognition
138
To create “… a professional community that discusses new teacher materials and strategies and that supports the risk taking and struggle
entailed in transforming practice.”
Teach Epidemiology
Teaching Epidemiology
139Teach Epidemiology
Teaching Epidemiology
Group
Assignments
Births: Class 1, p. 6-12
War: Qs 11-21
Case-control: Class 1, p. 16-21
Confounding: p. 32-36
Bias: p. 25-29 and 30-32
Alpine Fizz: Procs 2, 4, 5
top related