Epidemiology and Lessons from Humans

Question 1

Define biostatistics

Answer 1

Biostatistics is application of statistical methods in biology, medicine and public health

Question 2

Define Epidemiology

Answer 2

Epidemiology is the study of patterns ofhealthand illness and associated factors at thepopulationlevel

Question 3

Define an Observational based study

Answer 3

Studies that do not involve any intervention or experiment

Question 4

Define an Experimental study

Answer 4

Studies that entail manipulation of the study factor (exposure) and randomization of subjects to treatment (exposure) groups

Question 5

Define the Null Hypothesis

Answer 5

Also known as H0 - there is no association between a risk factor and the outcome

Question 6

Define the Alternative hypothesis

Answer 6

Also known as Ha - there is an association between a risk factor and the outcome

Question 7

Define and describe a Type 1 error

Answer 7

A false positive - error of rejecting a null hypothesis that is actually true
Pregnant woman isn’t pregnant but test says she is

Question 8

Define and describe a Type 2 error

Answer 8

False negative - error of failing to reject a false null hypothesis when it is false
Pregnant woman is pregnant but test say she isn’t

Question 9

Define a Case-Control

Answer 9

Selects based on disease status, then looks back at exposure

Question 10

Define a cohort study

Answer 10

Selection into study on basis of exposure status, then follow over time

Question 11

Define a confounding variable

Answer 11

extraneous variable that correlates with both the dependent variable (lung cancer) and the independent variable (yellow fingers)

Question 12

Explain how chance invalidates a study

Answer 12

Often due to inadequate sample size so that the hypothesis should not be rejected (Type II error)

Question 13

Explain how systematic Bias invalidates a study

Answer 13

A conclusion may be erroneous if it is Biased: that is, the conclusion is calculated in such a way that it is systematically different from the population parameter of interest

Question 14

State two forms of Bias

Answer 14

There are different sources of Bias, including Selection Bias (asking people on the street to answer a question) and Measurement Bias (such as the classic 1963 Rosenthal and Fode “maze-smart” rat experiment)

Question 15

Explain how confounding can invalidate a study

Answer 15

Example - Yellow Fingers cause Lung Cancer, so we must stop people from having Yellow Fingers!

Question 16

Explain what a randomised controlled trial is

Answer 16

Exposure is assigned randomly - basis on which groups are formed at beginning of the study is random

Question 17

State the 7 ethical requirements of a study

Answer 17

Social or scientific value
Fair subject selection
independant review
scientific validity
Favourable risk-benefit view
Informed consent
Respect for enrolled subjects

Question 18

Define Equipoise

Answer 18

That there is genuine uncertainty in the expert medical community over whether a treatment will be beneficial. This indicates that the answer is not known and needs to be known

Question 19

State the role of the Data safety monitoring Board

Answer 19

It monitors a trial to see if it should be stopped early - if one group is doing much better than the other

Question 20

Define what is Relative Risk

Answer 20

Compares which event is risk is risker

Question 21

Define Relative Risk reduction

Answer 21

?

Question 22

Define Odds Ratio

Answer 22

How well is Group A is doing in comparison by Group B

Question 23

Define Absolute risk difference

Answer 23

Risk of A - Risk of B

Question 24

Define number needed to treat

Answer 24

The value means for every x people you’d expect one case