Epidemiology and statistics

Question 1

When is a Z score considered to be low?

Answer 1

• Z score is the number of standard deviations above or below the mean
• 95% of values are within 2 standard deviations of the mean
• Z score

Question 2

What is sensitivity?

Answer 2

The number of patients with a positive test who have a disease divided by all patients who have the disease.
TP/TP + FN

Question 3

What is specificity?

Answer 3

TN/FP + TN

Question 4

What is positive predictive value?

Answer 4

likelihood that a patient with a positive test has the disease.

TP/TP + FP

Question 5

What is negative predictive value?

Answer 5

TN/FN + TN

Question 6

Describe the various observational and experimental study designs.

Answer 6

epidemiology notes

Question 7

What is type 1 versus type 2 error?

Answer 7

Type 1 error = false positive–>reject the null hypothesis, when the null is actually true. Usually 0.05 is set as the threshold for type 1 error
Type 2 error = false negative–>accept the null, when it isn’t true.

Question 8

What is power? What factors affect power?

Answer 8

• Power is the probability of not commiting a type 2 error
• Factors: sample size, effect size, alpha level
• Lower power is lower sample size, smaller effect size (which depends on difference between means and standard deviation)
• Bigger difference between means and bigger standard deviation will each lead to higher power
• I think that power is used for sample size calculation

Question 9

What are the advantages and disadvantages of case control study?

Answer 9

Premise: start off with the disease/outcome (case) and controls and looks backwards at exposure status

Advantage: good for rare diseases, inexpensive, can control for multiple confounders

Disadvantage: recall bias, poor selection of controls can introduce bias, cannot calculate relative risk