Ratios & Risk

Question 1

Odds ratio equation

Answer 1

Odds of exposure in cases / Odds of exposure in controls

Question 2

Odds equation

Answer 2

probability / (1 – probability)

Question 3

Why do we look at odds ratio from case-control study?

Answer 3

it is not possible to calculate the incidence of disease in the exposed and non-exposed individuals

Question 4

Odds ratio of 1

Answer 4

exposure is no more likely in cases than controls

Question 5

Odds ratio < 1

Answer 5

exposure is less likely in case group

Question 6

Odds ratio > 1

Answer 6

exposure is more likely in case group

Question 7

Point prevalence equation

Answer 7

No. cases at set point in time/ No. people at set point in time

Question 8

Relative risk

Answer 8

used as a measure of association between an exposure & disease

Question 9

Relative risk equation

Answer 9

Incidence in exposed group / Incidence in unexposed group

Question 10

Attributable risk

Answer 10

measure of exposure effect that indicates how much greater frequency of disease in exposed group is vs. unexposed, assuming relationship between exposure & disease is causal

Question 11

Attributable risk equation

Answer 11

Incidence in the Exposed – Incidence in the Unexposed

Question 12

Sample

Answer 12

relatively small number of observations/ patients from which we try to describe the whole population from which the sample has been taken

Question 13

Sampling variation

Answer 13

differences in samples from same population

Question 14

Normal distribution

Answer 14

a symmetrical distribution

Question 15

How to deal with confounders at the design stage of a study

Answer 15

Randomisation
Restriction
Matching

Question 16

Example of restriction to deal with confounders

Answer 16

include patients in a clinical trial only aged 18- 65 without pre-existing illness so results of trial are not confounded by different levels of age or morbidity

Question 17

Matching to deal with confounders

Answer 17

controls are selected to have a similar distribution of potentially confounding variables to the cases, e.g. “matched” for sex

Question 18

How to deal with confounders at the analysis stage of a study

Answer 18

Stratification
Standardisation
Regression

Question 19

Stratification to deal with confounders

Answer 19

risks are calculated separately for each category of confounding variable e.g. age groups

Question 20

Standardisation to deal with confounders

Answer 20

Used to produce a SMR

Question 21

Regression to deal with confounders

Answer 21

statistical modelling is used to control for 1 or many confounding variables

Question 22

Critical appraisal

Answer 22

systematically examining research evidence to assess its validity, results and relevance before using it to inform a decision.

Question 23

Summarise a paper 1st before critically appraising

Answer 23

Why did they do it?
What did they do?
What did they find?
What did they conclude?

Question 24

Critical appraisal, consider

Answer 24

Question
Design
Population
Methods
Analysis
Confounders
Bias
Ethics
Interpretation

Question 25

Want to know whether difference is due to chancer statistically significant?

Answer 25

Set up a null hypothesis

Aim to disprove using evidence

Question 26

If P< 0.05

Answer 26

There is a significant difference

Null hypothesis can be rejected