Week 4

Question 1

Marginal probability

Answer 1

Probability of the occurrence of a single event

P(A)

Question 2

Joint probability

Answer 2

Probability that two events will occur simultaneously

P(A & B)

Question 3

Conditional probability

Answer 3

Probability of one even given that another event has occurred
P( A I B)

Question 4

Diagnostic testing

Answer 4

• important for health care

- are not 100% accurate

Question 5

Sensitivity and Specificity
what they are, how you calculate them, what it looks like 2x2 table, how to interpret them, how to find the error (false negative and false positive rate)

Answer 5

SENSITIVITY
the probability that the diagnosis test is positive in patients who truly do have the disease
SN = TP/(TP+FN) x100 to get %

SPECIFICITY
the probability that the diagnostic test is negative in patients who truly do not have the disease
SP = TN/(TN+FP) x100 to get %
**want SN and SP to be as close to 1 as possible as that means 100%

ERROS IN DIAGNOSITC TESTING
false negative rate = 1 - sensitivity (test is negative but disease is positive
false positive rate = 1 - specificity (test is positive but disease is negative)
** we want these to be as close to zero which means that the diagnostic test is 100% accurate

in 2x2 table
TP FP
FN TN
first column used for calculating sensitivity
second column for calculating specificity

Question 6

What is Odds Ratio and how do you interpret it?

Answer 6

Odds ratio is the measure of association between an exposure and outcome
- it is the odds that an outcome will occur given a particular exposure compared to the odds that the outcome occurs without exposure

INTEPRETATION
1 = no association
>1 = positive association - the exposure increased the odds of the event occurring
<1 = negative association - the exposure decreased the odds of event occurring

Question 7

What is Risk Ratio and how do you interpret it?

Answer 7

Risk ratio is the measure of the risk of an event occurring in both groups, exposure and don’t have exposure (people having lung cancer whether they drink alcohol or don’t. With relative risk you can better establish causality than with odds ratio.

INTEPRETATION
1 = risk is the same
>1 = increased risk of event occurring
<1 = decreased risk of event occurring

Question 8

What is Risk difference and how do you interpret it?

Answer 8

Risk difference is the absolute measure which gives you the absolute difference between the risks in each group

INTREPRETATION
minus value = exposure is harmful to the outcome
positive value = exposure is beneficial to the outcome
0 value = no difference between groups

if risk difference is 0.11 - this is a positive value therefore there is 11% less risk of liver cancer for non drinkers when compared to drinkers.

Question 9

What is Number Needed to Treat and how do you interpret it?

Answer 9

Number needed to treat is the amount of people who need to receive the intervention for one person (who also had the intervention) to experience or avoid the outcome.
ie. for every 9 people who stopped drinking alcohol, 1 case of liver cancer can be prevented.

IDEAL IS 1! therefore everyone would stop drinking and no would get cancer.