causation

Question 1

recall steps for ebvm

Answer 1

1) formula an answerable question
2) search for evidence
3) critically appraise the evidence: study design, epidemiological measures, bias and confounding, causality
4) apply answer to patient
5) audit outcome

focusing on causality

Question 2

the child and the light switch

Answer 2

child concludes flicking switch causes light; tendency for simple one factor explanation of causation

in reality single cause outcomes tend to be exception; complex interplay of factors

Question 3

cause

Answer 3

event, condition or characteristic which plays an essential role in producing an outcome (ex occurrence of disease)

Question 4

causation

Answer 4

describes a combo of events that in correct sequence an timing inevitably result in an outcome

Question 5

disease control does not require dependency on exact knowledge of agent or pathogenesis; give an example

Answer 5

cholera epidemic in london 1854 john snow removed handle of water pump, prevented disease without knowing infectious agent

Question 6

inductive reasoning

Answer 6

• based on repeated observations, patterns, formulation of hypothesis to explore and finally general conclusions
• generalize from observations and develop scientific laws

example: edward jenner; milkmaid who develop cowpox don’t get smallpox; vaccine invented

Question 7

deductive reasoning

Answer 7

• review what is already known, formulate hypothesis, collect data to test NULL hypothesis, lead to confirmation or not of original theory
• use general theory to predict cases

Question 8

3 causal models

Answer 8

1) host- agent- environment triangle
2) component cause model
3) caudal diagrams

Question 9

host agent environment triangle

Answer 9

• factors that affect risk of disease put into either host, egnt or environment factors

Question 10

limitations of host agent environment triangle

Answer 10

• does not account for timing of sequence of events
• does not show how factors may be interlinked
• overemphasis on agent factors (not appropriate for toxins/ non infectious agents)

Question 11

component cause model

Answer 11

• how many pieces needed to have full pie; when pie is full disease occurs
• Component causes; each piece of the pie, equal partners in producing effect
• Sufficient causes; represents the whole pie; set of components that in combo is producing the disease (particular disease might be produced by different sufficient causes- different pies)
• Necessary causes; the most important piece of the pie; factors that must be present for disease to occur
• Often not necessary to identify components (all pie pieces)
  Some disease have no necessary cause (no most important piece of pie)
• Necessary cause if often not sufficient cause (most important piece of pie does not equal the whole pie, other components are needed)
• Not all components must act at the same time (can interact years apart)

Question 12

in component cause model what are component causes

Answer 12

each piece of the pie, equal partners in producing effect

Question 12

in component cause model what is sufficient causes

Answer 12

represents the whole pie; set of components that in combo is producing the disease (particular disease might be produced by different sufficient causes- different pies)

Question 13

in component causal model what is necessary causes

Answer 13

the most important piece of the pie; factors that must be present for disease to occur

Question 14

causal diagrams

Answer 14

• path digram
• complex web of interactions

Question 15

give 2 ways causal diagrams can be used

Answer 15

• to plan disease control activities
• to perform some types of statistical analyses

Question 16

causal direct; direct and indirect causes/ exposures

Answer 16

direct; no known intervening variable between exposure and disease

indirect- effect of exposure if mediated through on or more intervening variables

Question 17

3 parts to causal diagram

Answer 17

• outcome/ effect (disease)
• causes/ exposures (direct or indirect) can be both direct and indirect
• direction of associated (linear ie pos, neg, or non linear)

Question 18

linear and non linear causes

Answer 18

• linear; pos or negative
• non linear; curve ex lambs with low and high birth weight increased risk of death

Question 19

advantages of causal diagram

Answer 19

• possible to specify the sequence of effects
• possible to specify interrelationships between factors
• effects of any gents are more likely to be seen within context w other factors

Question 20

3 causal guidelines

Answer 20

• koch’s postulates
• hill’s criteria
• evan’s concept of causation

Question 21

koch’s postulates

Answer 21

conditions required for an agent to be considered as cause of a disease

1) agent must be present in every case of disease
2) agent has to be isolated and grown in pure culture
3) agent must cause specific disease
4) once isolated agent must reproduce disease and must be recovered from experimentally infected animals

Question 22

what was first disease to meet koch’s postulates

Answer 22

anthrax

Question 23

problems with koch’s postulates

Answer 23

ignored environmental influence, host factors, mixed infections and non infectious diseases

Question 24

hill's criteria

Answer 24

9 criteria that contribute a different amount of strength to likelihood a relationship between potential risk factor and disease is causal 1) strength 2) consistency 3) specificity 4) temporality 5) biological gradient 6) plausibility 7) coherence 8) experiment 9) analogy

Question 25

hill's criteria 1) strength of association

Answer 25

- strong associations more likely to be causal - indicated by risk ratio greater than 2 - weak association does NOT rule out causal relationship

Question 26

hill's criteria 2) consistency

Answer 26

- consistent findings in studies using different methods and diff populations - lack of consistency does NOT rule out a causal relationship

Question 27

hill's criteria 3) specificity

Answer 27

- when single cause produces a specific disease - many exceptions - when present, specificity does provide evidence of causality but its absence doe not rule out causation

Question 28

hill's criteria 4) temporality

Answer 28

- cause must be before effect - if A comes after B then A did not cause B - can be difficult to confirm

Question 29

hill's criteria 5) biological gradient

Answer 29

- dose-response relationship - greater exposure should result in greater effect - non linear effects may be present - absence of dose-response does NOT rule out causal relationship

Question 30

hill's criteria 6) plausibility

Answer 30

- association makes biological sense; agrees w currently accepted understanding of processes - not objective

Question 31

hill's criteria 7) coherence

Answer 31

- relates to absence of conflicting info - fine distinction from plausibility

Question 32

hill's criteria 8) experiment

Answer 32

- investigator initiated interventions that modify exposure through prevention, treatment or removal should result in less disease - randomized control trials best

Question 33

hill's criteria 9) analogy

Answer 33

has a similar relationship been observed with another exposure and or disease

Question 34

evan's concept 10 rules

Answer 34

- set out criteria for judging whether or not exposures cause disease 1) the proportion of individuals with disease should be higher in those exposed than those not 2) exposure to cause should be more common in those with disease than without 3) number of new cases should be higher in those exposed than those not 4) disease should follow supposed cause 5) a spectrum of host responses from mild to severe should follow exposure 6) host response should appear following exposure in those exposed but not in those not exposed 7) experimental reproduction should occur more often int hose exposed than those not exposed 8) elimination of cause should decrease frequency of occurrence 9) prevention/ modification of the host response should decrease the disease 10) all relationships should be biologically and epidemiologically credible