Measures of disease association and effect

Question 1

WHAT ARE MEASURES OF EFFECT?

Answer 1

measures of effect are constructs that quantify the association between effect and outcome.
-is the exposure associated/related to the outcome?
-if yes, what is the strength of the association?

Question 2

what is surveillance time?

Answer 2

the total time per 1000 person years to reach the end point for all participants within each group at risk of the end point

Question 3

What are the types of measures of exposure effect?

Answer 3

1. relative = ratios
   - Divide occurrence of outcome in exposed and unexposed
2. absolute = differences
   - Subtract the occurrence of outcome in exposed and unexposed

Question 4

what are the measures of relative exposure effect?

Answer 4

1. risk ratio/ relative risk
2. incidence rate ratio
3. odds ratio
4. prevalence ratio
   (in these measures. 1 = no association, values can be 0 to infinity, there are no units)

Question 5

1. in what type of study is odd’s ratio most commonly used, and why?
2. what other studies can it be used in?

Answer 5

1. case control studies because you do not know the incidence.
2. cohort
3. intervention
4. cross sectional

Question 6

in what type of study is prevalence ratio most commonly used? why?

Answer 6

cross sectional because it cannot calculate relative risk.

Question 7

what is the interpretation of all ratios?

Answer 7

for cumulative incidence (RR), incidence density (IRR) ,odds, and prevalence (and other ratios)
- 1 = null effect = no association, equal in exposed and unexposed
- >1 = positive assoc = outcome is higher exposed group
- <1 - negative assoc = outcome is lower in exposed group

Question 8

what are the measures of absolute exposure effect?

Answer 8

1. cumulative incidence difference
2. incidence rate difference
   (range is -infinity to +infinity , 0 means no association, same units as CI and IR)

Question 9

What is the number needed to treat? NNT

Answer 9

Number of people needed to be treated with the intervention to prevent one event, than if those same people had been treated with the comparison (e.g., placebo)

Question 10

what is NNT called if it is a positive number?

Answer 10

number needed to harm

Question 11

what does prevalence tell us?

Answer 11

Prevalence tells us how many cases of a disease (or an outcome) there are in a given population at a specified time – how many people have the outcome

Question 12

What does incidence tell us?

Answer 12

Incidence tells us how many new cases appear in a population over a specified time period – how many people will get the outcome

Question 13

identify which measures of exposure effect are calculated in different study designs.
1. cross sectional
2. cohort
3. case control
4. intervention

Answer 13

1. cross sectional = odds + prevelance
2. cohort = odds + relative risk + risk difference
3. case control = odds
4. intervention = odds + relative risk + risk difference