Measures of Association

Question 1

What is the KEY to epidemiology?

Answer 1

COMPARISON

Question 2

What is a hypothesis?

Answer 2

developed from your question

predictive statement

relationship between independent and dependent variables

Question 3

What is an independent variable?

Answer 3

characteristic or event that is hypothesized to cause or contribute to the variation or other variables

ie- exposure/risk factors

Question 4

What are dependent variables?

Answer 4

value depends on the other variables-

the outcomes, the disease

Question 5

What is the Null Hypothesis?

Answer 5

No relationship between the independent and dependent variables

Question 6

What is the alternate hypothesis?

Answer 6

there is a relationship between the independent and dependent variables

Question 7

What are ratio measures of Association?

Answer 7

asses the strength or magnitude of the statistical association between the exposure and disease of interest

in cohort studies use relative risk (rick ratio)

in case-control studies use odds ratio

Question 8

What is risk

Answer 8

the probability of a disease-free individual developing given disease over a specified period, conditional on that individual not dying from any other disease during the period

risk is w/o units from 0 to 1

risk = attack rate in outbreak settings, incidence proportion, cumulative incidence

number of events/population at risk at start of period

Question 9

What are odds?

Answer 9

probability of an event occurring compared to the prob of it not occurring

prob event occurs/prob event does not occur

Question 10

In cohort studies use __________

in case studies use __________

Answer 10

cohort use odds ratio

case studies use risk

Question 11

What is relative risk?

Answer 11

estimates the magnitude of an association, between exposure and disease

relative risk is the excess risk in the exposed group compared to the unexposed (background expected group)

RR indicates the likelihood of developing a disease for the exposure relative to the unexposed

Expressed as a ratio

RR = Risk exposed / Risk unexposed

Question 12

How do we interpret Relative Risk?

Answer 12

those who are exposed have RR times the risk of developing the dz as those who were unexposed

the risk of dz among the exposed is RR times higher /lower than the risk of dz among the

% increased/decrease tisk

Question 13

What does RR >1.0 mean?

Answer 13

risk is greater in the exposed group than the unexposed group

Question 14

What does RR

Answer 14

the exposed group is less likely to have the dz than the unexposed group

Question 15

Why do we use odds ratio in case-control studies?

Answer 15

you cannot directly calculate the risk of disease because you do not have a denominator for the population

you can calculate the odds of exposure among cases and controls

the OR compares the odds of exposure among cases to the odds of exposure among controls

Question 16

How is the Odds Ratio related to the 2x2 table?

Answer 16

it reduces to AD/BC the cross product ratio in a 2x2 table

Question 17

How do we interpret odds ratio?

Answer 17

those with disease were OR times as likely to have the exposure compared to those w/o disease

odds of exposure among the diseased is OR times higher/lower than the odds of exposure among the non-diseased

Question 18

What do the actual numbers mean?

Answer 18

0- 0.3 strong benefit

0.9-1.1 no effect

> 2.5 strong effect

Question 19

compare odds ratio vs relative risk

Answer 19

OR is a good estimate of RR when a disease is infrequent

OR is not a good estimate of RR when a disease in NOT infrequent

Question 20

What are measures of potential impact?

Answer 20

reflect the apparent contribution of an exposure to the frequency of disease in a population

important for policy makers and funding sources to understand impact a prevention program might have

Question 21

What is risk difference?

Answer 21

aka attributable risk

risk diff = risk exposed - risk unexposed

Risk diff = incidence exposed - incidence unexposed

excess risk in those exposed

Question 22

What is attributable risk percent?

Answer 22

only appropriate if RR >1

proportion of cases in the exposed group attributable to exposure

the most that we can hope to accomplish in reducing risk of disease if we completely eliminate the exposure

(Risk exposed - risk unexposed) / Risk exposed

Question 23

What is the population attributable risk percentage?

Answer 23

proportion of cases in the entire population (exposed and unexposed) attributable to exposure

(Risk pop - Risk unexposed) / Risk population

Question 24

What does attributable risk assume?

Answer 24

ASSUMES the association is causal

Question 25

What is causation?

Answer 25

the association between exposure and the outcome is real does exposure cause disease

Question 26

What are the guidelines for assessing causality?

Answer 26

temporal relationship strength of association dose-response relationship replication of findings biological plausibility consideration of alternate explanations cessation of exposure consistency with other knowledge specificity of assocation

Question 27

Temporal relationship

Answer 27

if a factor is believed to be the cause of a disease, the exposure to that factor must occur before disease develops

Question 28

Strength of association

Answer 28

measured by the relative risk or odds ratio stronger association = more likely the relationship is causal

Question 29

Dose- response relationship

Answer 29

as the dose of exposure increases, risk of disease also increases if d-r present, it is strong evidence of a causal relationship however the absence of a dose-response relationship, does not rule out a causal relationship

Question 30

Replication of findings

Answer 30

expect to see the same findings in different studies and in different popualtions

Question 31

Biological plausability

Answer 31

the findings should be consistent with biologic knowledge

Question 32

Consideration of alternate explanations

Answer 32

important to report that alternative explanations (bias or confounding) have been considered

Question 33

Cessation of exposure

Answer 33

if a factor is a cause of a disease we would expect the risk of disease to decline when exposure to the factor is reduced or eliminated

Question 34

Consistency w/ other knowledge

Answer 34

causal relationships have findings consistent w/ other data

Question 35

Specificity of association

Answer 35

when a certain exposure is associated w/ the only one disease`

Question 36

How do we assess explanations for an association

Answer 36

if we determine an association between an exposure and an outcome exists, it might be because: bias, confounding, error in conducting the study, due to chance

Question 37

What is bias?

Answer 37

systematic error in design, conduct or analysis of a study that results in a mistaken estimate of an exposure's effect on the risk of disease

Question 38

What are the 2 major categories?

Answer 38

selection bias, information bais

Question 39

What is selection bais?

Answer 39

occurs whenever identification of individual subjects for an inclusion results in a mistaken estimate for the measure of the effect prob with who is in the study ex- detection, self-selection, non-response, inappropriate comparison group

Question 40

What is information bais?

Answer 40

systematic error in the collection of exposure outcome data that results in a mistaken estimate of an exposures effect on risk of disease problem with the info you collect ``` bad questions interview bias respndent errors record bias misclassification bias social desirability bias ```

Question 41

define confounding

Answer 41

distortion of an exposure-disease association by the effect of some third factor- the confounder both exposure-confounder and disease-confounder interactions results in a mistaken estimate of an exposure's effect on the risk of disease can be eliminated during data analysis

Question 42

What are the requirements to be a confounder?

Answer 42

must be: associated with the outcome associated with the exposure, but is not a result of exposure