Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

EVIDENCE BASED MEDICINE - 2 > 13 - Causation > Flashcards

13 - Causation Flashcards

1
Q

What is the basic definition of cause?

A
  • Any factor that produces a change in the severity or frequency of the outcome
  • *do NOT need to understand ALL causal factors to prevent or at least control disease
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is inductive reasoning?

A
  • Process of making generalized inferences about causation based on repeated observations
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Inductivism and logical fallacies: example with the rooster

A
  • Rooster crows just before sun rise
  • Therefore, roosters crowing causes the sun to rise
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Koch’s postulates: limitations

A
  • IGNORES environmental factors
  • NOT applicable to non-infectious diseases
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Epidemiology vs. the lab

A
  • Can’t always recreate disease in lab
  • If wanting to understand complex issues affecting disease in a natural world then need to study the NATURAL WORLD
  • *need both natural world study and lab studies
  • *most causation discussion are LIMITED to observational research rather than experimental
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Observational vs. experimental research

A
  • Observational: looking for cause
  • Experimental: looking for effects
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Experimental studies

A
  • We RANDOMIZE individuals to receive a factor and some to receive nothing
  • We know the factor precedes disease and other variables accounted for by randomization
  • We contrast outcomes in treatment and control
  • Assume EXCHANGEABILITY
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Observational studies

A
  • Estimate outcome differences between individuals that happen to vary in their exposure status
  • Matching and restriction where appropriate to minimize differences between groups
  • *measure ASSOCIATION between changes in exposure and outcome
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What are the limits to experimental studies?

A
  • Difficult to duplicate realistic dose, exposure pathway or complete set of typical cofactors
  • Difficult to carry out experiments that actually resemble “real-world” conditions
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Observational comparisons: what are you comparing it to?

A
  • Ex. compare to current treatment (can’t just have totally untreated animals)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Cohort studies: 2 steps

A
  1. Define groups (cohorts) of animals according to exposure of animals in groups to factors of interest
  2. Follow groups FORWARD IN TIME to see which animals develop the disease under investigation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What do you compare with cohort studies?

A
  • Risk in exposed and unexposed groups
  • *reported as RELATIVE RISK
  • Can look at more than one disease resulting form a specific type of exposure
  • **CLOSEST OBSERVATIONAL STUDY WE CAN GET TO RCCT
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Case-control studies: 2 steps

A
  1. Define groups of diseased and healthy animals
  2. Assess whether animals in the 2 groups have differences in past exposure to different risk factors
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What do you calculate in case-control studies?

A
  • ODDS RATIO to indirectly estimate RR provided that incidence of disease is low and cases + controls are truly random samples from the same population
  • Good for studying RARE DISEASES
  • Can assess more than one exposure in the same study
  • *watch for recall bias (did exposure actually come before the disease)
    o Hard when there is a long latent period (Ex. cancer)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Statistically significance does NOT equal causality

A
  • To prove causal association we need to describe a chain of events
    o From cause to effect at the molecular level
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

**What is confounding or a confounder?

A
  • Effect of an extraneous variable that can wholly or partly account for an apparent association between variables in an investigation
  • *can produce a spurious association between study variables, or can mask a real association
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What are the 3 ‘criteria’s’ that a confounder must be?

A
  1. Be associated with the response variable
  2. Be associated with risk factor (exposure or treatment) of interest
  3. Not be an intervening or intermediate step between the risk factor and response
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Component model of causation

A
  • ALL disease is MULTIFACTORIAL
  • Sufficient vs. necessary causes
  • Casual mechanism remains constant
  • *strength of association between exposure of interest and outcome will VARY
    o *depends on distribution of risk factors
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What is a sufficient cause?

A
  • if it inevitably produces an effect
    o Virtually ALWAYS comprises a number of COMPONENT CAUSES
  • Particular disease may be produced by different sufficient causes
20
Q

What is a necessary cause?

A
  • If a risk factor is a component of EVERY SUFFICIENT CAUSE
21
Q

What are the components of a sufficient cause?

A
  • Factors may present concomitantly or may follow one another in a chain of events
  • When there are a number of chains with one or more factors in common then we have a ‘causal web’
22
Q

What is causal complement?

A
  • The SHARED COMPONENT CAUSES that make up a sufficient cause
23
Q

Interaction among causes

A
  • 2 or more component causes acting in the SAME SUFFICIENT CAUSES INTERACT CAUSALLY TO PRODUCE DISEASE
24
Q

What is the objective of epidemiological investigations of cause?

A
  • The ID of sufficient causes and their component causes
  • *removal of one or more components from a sufficient cause swill then PREVENT disease produced by the sufficient cause
25
Q

What is a web of causation?

A
  • Direct and indirect causes representing a chain of actions with indirect causes activating direct causes
26
Q

Relationships are shown using a causal diagram

A
  • Direct causes are often the PROXIMAL causes emphasized in therapy
  • Indirect causes are where effects of exposure are mediated through one or more intervening variables
27
Q

***What are the Hill’s criteria for causality?

A
  1. Temporality
  2. Strength of association
  3. Biological gradient or dose response
  4. Coherence or plausibility
  5. Consistency
  6. Specificity
  7. Analogy
  8. Experimental evidence
28
Q

**Hill’s criteria: time sequence

A
  • Cause must ALWAYS PRECEDE effect in time
    o *but same factor could occur again after disease in some individuals
  • *difficult to establish time sequence, especially with surrogate exposure measures
29
Q

Hill’s criteria: strength of association

A
  • Strong statistically significant association between factor and disease INCREASES likelihood that the factor is causal
  • Assumes that it is less likely that residual confounding could explain the result
30
Q

What does strength of association depend on?

A
  • Distribution of other components of the sufficient cause
31
Q

What is an example of an important weak association that has been consider causal?

A
  • Environmental tobacco smoke and lung cancer
32
Q

What is an example of a strong association due to confounding?

A
  • Birth order and Down’s syndrome
33
Q

Hill’s criteria: biological gradient

A
  • Dose-response relationship between a factor and disease INCREASES PLAUSIBILITY of factor being causal
  • *exceptions to linear change: threshold
  • Most should have a gradient that never changes direction (monotonic)
  • *alcohol consumption and death=J-shaped curve
34
Q

Hill’s criteria: consistency

A
  • Repeated observations of an association in different populations under different circumstances
  • Associations can be causal under unusual circumstances
  • *statistical significance should NOT be used to assess consistency
  • *example systematic reviews and meta-analysis
35
Q

Hill’s criteria: coherence/plausibility

A
  • Compatibility with existing knowledge
  • More reasonable to infer that a factor causes a disease if a plausible biological mechanism has been IDed than if such a mechanism is NOT known
36
Q

Hill’s criteria: specificity

A
  • A cause can lead to a single effect OR an effect has one cause
  • *not necessary, but can be supportive when it can be logically deduced from the causal hypothesis
37
Q

Hill’s criteria: analogy

A
  • Too subjective and open
  • Possible source of new hypothesis
  • Ex. smoking and stomach cancer plausible because smoking has been associated with a number of other cancers
38
Q

Hill’s criteria: experimental evidence

A
  • Clinical trials, animal lab experiments OR both
  • Uncertainty in extrapolating across species and outside tested dose ranges
39
Q

Hill’s criteria has reservations and exceptions

A
  • They are ‘viewpoints’ or ‘perspectives’
  • *there is significant individuality in interpreting the same evidence
    o Only agreed 68% of the time
40
Q

There must be a MATHEMATICAL ASSOCIATION between exposure and hypothesized effect or outcome

A
  • In most cases, outcome (disease) have a monotonic association with the increasing exposure
  • *besides temporality, there are NO criteria that are either necessary or sufficient
41
Q

What are some sources of area?

A
  • Chance
  • Systematic error in design of study or data
  • Confounding or mixing effects
  • Consistency
  • *observed association must NOT be ENTIRELY due to any source of error
42
Q

How can chance be assessed?

A
  • Correct application of statistical analysis (ex. p-values or CI)
43
Q

Systematic error in the design of study or data

A
  • Sampling bias
  • Misclassification of exposure or disease status
44
Q

Confounding or mixing effects

A
  • Resulting from a third unaccounted for risk factor associated with BOTH exposure and disease
45
Q

Consistency

A
  • Across different study designs and different study populations is SUPPORTIVE

EVIDENCE BASED MEDICINE - 2 (6 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions