Epi

Question 1

How does prevalence affect PPV/NPV/Se/Sp

Answer 1

Se and Sp don’t change.

Increase in Prevalence = Increase PPV, Decrease NPV

Question 2

Cross sectional and cohort study for incidence and prevalence

Answer 2

Cross sectional: Prevalence

Cohort: Incidence

Question 3

If there is skew in a distribution curve, how are the measures of central tendency affected

Answer 3

Mean-Median-Mode

Mean moves the move, Mode stays in the middle

Question 4

Test to compare means of 2 groups

Answer 4

t-test (not Chi). If you’re mean, have to take a timed test.

Question 5

test to compare categorical data (% or proportions) of 2 or more groups

Answer 5

Chi square. Does NOT compare mean values.

“percentage of members of 3 different ethnic groups who have hypertension)

Question 6

Test to compare 3 or more means

Answer 6

ANOVA (t-test to compare 2 means)

Question 7

what kind of bias should you think about when you see a new screening test for poor prognosis dz like lung cancer or breast cancer?

Answer 7

Lead-time bias

apparent survival time increases only b/c dz is detected earlier because youre screening for it

Question 8

T/F: Precision can be reflected by confidence interval

Answer 8

true…narrower CI = more precise

Question 9

difference between case-control and retrospective cohort

Answer 9

CCS: Outcome first, then determine risk factors
Retro: first ascertain risk factor exposure, then assess outcomes

Question 10

significant loss to follow up causes what kind of bias

Answer 10

Attrition bias, a form of SELECTION bias