Stats Flashcards

1
Q

How many values lie within 1 2 and 3 SD of the mean

A
  1. 3%
  2. 4%
  3. 7%
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Two hundred patients with asthma are recruited. Of these 200, 40 report either one or both parents smoking in the house when they were a child. A further 200 controls who do not have asthma are recruited. Of these people 20 report that one or both parents smoked in the house. What is the odds ratio of asthmatics having been exposed to passive smoking as a child?

A

2.25

Odds of 40:160 = 1:4
Vs odds of 20:180 = 1:9

Odds ratio is (1/4)/(1/9)
=2.25

Odds is ratio of the number of people who incur a particular outcome to the number of people who do not incur the outcome

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Calculate number needed to treat:

Percentage of patients having further MI within 3 months
Aspirin 4%
Aspirin + new drug 3%

A

1/(Absolute risk reduction)

so 1/(0.04-0.03) = 100

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

what is the purpose of a funnel plot?

A

a funnel plot shows publication bias in meta-analyses - symmetrical means bias unlikely

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is hazard ratio

A

Similar to relative risk but used when risk is not constant to time… Ie survival over time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What best describes the characteristics of a negatively skewed distribution?

A

Alphabetical order so
Mean median mode
Than > is positive and< is negative

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is linear regression?

A

linear regression may be used to predict how much one variable changes when a second variable is changed

Correlation is used to test if there is an association in the first place

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Calculate absolute risk reduction for group of 40 patients, 2 had particular outcome and out of 40 control 1 had outcome

A

1/20 - 1/40

= 0.025

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the power of a test

A

= 1 - the probability of making a type 2 error

Which is accepting the null hypothesis when it’s false

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Which type of study design is retrospective?

A

Case-controlled study

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Which is related to prevalence - sensitivity or positive predictive value

A

PPV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Examples of parametric test

A

Student t test (one group or two groups)
Pearson (to find correlation)
ANOVA (if more than 2 means compared)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Skewed distribution negative and positive

A

alphabetical order: mean - median - mode

‘>’ for positive, ‘

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What statistical test do you use for Parametric data, comparing two groups

A

Student t test unpaired

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Clinical study phases 1

A

Pharmacokinetics /dynamics

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Clinical study phase 2

A

2a assesses dosing

2b assesses efficacy

17
Q

Clinical study phase 3

A

Assesses effectiveness

18
Q

Clinical study phase 4

A

Long-term post marketinf surveillance and side effects

19
Q

Calculate Absolute risk reduction

A

Control event rate - experimental event rate

Remember rate, i.e. number who had outcome/total

20
Q

Calculate NNT

A

NNT = 1 / absolute risk reduction

= 1 / ( control rate - experimental rate)

21
Q

Which type of statistical measure is it most appropriate to compare survival time with?

A

Hazard ratio

  • similar to relative risk but is used when risk is not constant to time. It is typically used when analysing survival over time
22
Q

What is the standard error of the mean?

A

= Standard deviation / sq root of number of patients

23
Q

What percentage of adults will have an IgE level higher than 2 standard deviations from the mean?

A

2 SD = 95.4%

100-95.4 = 4.6%

There are the two extremes, so just for patients higher is 2.3%

24
Q

Type 1 error Vs type 2 error

A

Type 1 the null hypothesis is rejected when it is true

Type 2 null hypothesis is accepted when it is false

25
Q

What level of evidence does a study offer which is obtained from a meta-analysis of randomised controlled trials?

A

1a

26
Q

What is equivalent to the pre-test probability?

A

Prevalence

27
Q

What is variance

A

Variance = square of standard deviation

28
Q

Odds of rolling a 6 with dice

A

1/5 or 0.2

Probability is 1/6