Dit - Epidemiology And Ethics

Front 1

HYQ: Assuming a normal bell shapped distribution, what percentage of the study population falls within 1 std, 2 std, and 3 stds?

Back 1

1: 68%
2: 95%
3. 99.7%

Front 2

What is the Eq for Odds Ratio and Relative Risk?

AB
CD

Back 2

OR= AD/BC
RR= (A/A+B)/(C/C+D)

Front 3

HYQ: when is the odds ratio a good approximation of the relative risk?

Back 3

Low prevalence

Front 4

HYQ: What does it mean when the Relative Risk =1?

Back 4

there is no relationship btw the risk factor and disease

Front 5

What should you do if parents refuse life saving care for their child?

Back 5

In an emergency --> treat now and seek a court order

In a non-emergency --> seek a court order

Front 6

HYQ: under what circumstances are you allowed to breach confidentiality?

Back 6

If the pt says so
If the pt is suicidal or homocidal
If the pt has been involved in a penetrating assult (knife or gun shot)
If the pt has a reportable dz

Front 7

HYQ: in what situations does an unemancipated minor not require parental consent to treatment?

Back 7

Pregnancy
Birth control pills
STD treatment
Drug rehab

Front 8

What is type I error?

Back 8

alpha - the study says there is a relationship when none exists

Falsely rejecting the null

Front 9

What is type II error?

Back 9

Beta - the study says NO relationship when one exists

Falsely accepting the null

Front 10

What is the Eq for sensitivity?

AB
CD

Back 10

A/A+C

Front 11

What is the Eq for Specificity?

AB
CD

Back 11

B/ (B+D)

Front 12

What is the Eq for PPV?

AB
CD

Back 12

A/(A+B)

Front 13

What is the Eq for NPV?

AB
CD

Back 13

D/(D+C)

Front 14

How do you calculate the Attributable Risk?

Back 14

The % of at risk with dz - the % of not at risk with the dz

For ex: 5% of smokers have pneumonia, and 1% of non-smokers have pneumonia... the AR of smoking is 4%

Front 15

What is the NNT?

Back 15

Number Needed to Treat

NNT = 1/ Absolute risk reduction

For ex: if 10,000 pts took t-pa during a STEMI and 100 were saved, the NNT would be (10,000/100) = 100

Front 16

What does it mean if the CI for the Odds Ratio crosses 1?

Back 16

there is no association

Front 17

What does it mean if the CI for 2 treatment groups crosses 0?

Back 17

there is no diff btw treatments

Front 18

What type of trial is a prospective comparison of effectiveness?

Back 18

RCT- the gold standard b.c it is double blinded and randomized

Front 19

What type of study is prospective,examines a group of subjects exposed to a given factor, and calcuates the Relative Risk?

Back 19

Cohort- RR, good for rare exposures

There is a relative risk when hooking up with a coWhore

Front 20

What type of study is retrospective, good for rare diseases and measures Odds Ratio?

Back 20

Case-Control study

"chicken or fish, case or control"

Front 21

What type of study uses a survey?

Back 21

Cross-sectional