Other

Question 1

Sources of morbidity data

Answer 1

disease reporting, insurance companies; hospital & healthcare records

Question 2

Motality can be a proxy measure for incidence when

Answer 2

1. case-fatality is high

2. duration of disease is short

Question 3

Mortality data is the index of

Answer 3

1. severity of disease in a population (force)
2. overall health status of a population
3. effectiveness of public health services, preventive medicine, and therapy.

Question 4

How to calculate PAR%

Answer 4

Pe(RR-1)/(Pe(RR-1)+1).
e.g. In the defined pop, 8% of CHD occurrence is attributed to smoking and thus might be prevented if smoking were eliminated from the pop.

Question 5

CI Provide?

Answer 5

1. proxy of statistical power

2. measure of precision for observed estimate (reliability)

Question 6

How to calculate CI

Answer 6

ln(OR/RR) +- 1.96*SE

then ex..

Question 7

p-value for trend

Answer 7

supports whether there’s a dose-response pattern. <0.05, then it’s does-response.

Question 8

ecological study

Answer 8

aggregated group data.

preliminary evidence for confirmation by individual level study

Question 9

Ecological fallacy

Answer 9

inappropriate conclusions about individual-level relationships based on aggregate data.

Question 10

cross-sectional

Answer 10

exposure and disease assessed at the same time.
generate hypothesis for etiologic study
monitor person, place, time trends

Question 11

prevalence odds ratio

Answer 11

odds of having disease. in cross-sectional study

Question 12

how to make a cohort study out of cross-sectional

Answer 12

exclude prevalent cases, implement follow-up

Question 13

T/F: controls should be representative of the base population that gave rise to cases

Answer 13

T

Question 14

Sources of control

Answer 14

1. hospital
2. community based
3. neighborhood
4. special groups (friends, co-worker)

Question 15

advantages of matching

Answer 15

1. enhances study efficiency

2. can provide control for confounding

Question 16

disadvantages of matching

Answer 16

1. matching on several variables increases difficulty finding controls
2. cannot examine association with matching variables
3. make cases and controls more similar on exposure of interest. (residual confounding; matching on factors weakly associated with exposure can reduce statistical power)

Question 17

matched OR

Answer 17

B/C. after matching on….

Question 18

ad and disad of case-control

Answer 18

1. efficient, rare disease, multiple exposures with single disease.
2. temporality, no direct measure of incidence, rare exposure, selection bias and recall bias

Question 19

select sample in cohort

Answer 19

1. select on a characteristics not related to exposure (residence, occupation-nurse health study
2. select on exposure status - occupational exposures

Question 20

ways to select non-exposed

Answer 20

1. select from within a defined study population

2. select an external comparison group (general population - healthy worker effect)