Statistics

Question 1

state the calculation for power in a study

Answer 1

power = 1 - probability of type II error

type II error is when the study does not reject H0

> increasing sample size increases power and decreases type II error

Question 2

what is the equation used to calculate sensitivity

Answer 2

true positives / (false negatives + true positives)

Question 3

what is the equation used to calculate specificity

Answer 3

true negatives / (true negatives + false positives)

Question 4

what is the equation used to calculate positive predictive value

Answer 4

true positives / (true positives + false positives)

Question 5

what is the equation used to calculate negative predictive value

Answer 5

true negatives / (true negatives + false negatives)

Question 6

what is the equation used to calculate likelihood ratio for a positive result

Answer 6

sensitivity / (1 - specificity)

Question 7

what is the equation used to calculate likelihood ratio for a negative result

Answer 7

(1- sensitivity) / specificity

Question 8

how do you calculate number needed to treat (NNT)?

Answer 8

1 / absolute risk reduction

absolute risk reduction = control event rate (CER) - experimental event rate (EER)

Question 9

how do you calculate relative risk reduction>

Answer 9

(EER - CER) / CER

Question 10

state the coefficients used for
- parametric data
- non-parametric data

Answer 10

• parametric (normally distributed): Pearson’s coefficient
• non-parametric: Spearman’s coefficient

Question 11

describe a type I error

Answer 11

the null hypothesis is rejected when it is true

Question 11

describe the following non-parametric tests
- Mann Whitney U test
- Wilcoxon signed-rank test
- Chi-squared test

Answer 11

• Mann Whitney U test
  > compares ordinal, interval or ratio scales of unpaired data
• Wilcoxon signed-rank test
  > compares two sets of observations on a single sample e.g. before after test on same population following an intervention
• Chi-squared test
  > used to compare proportions or percentages e.g. compares the % of patients who improved following two different interventions

*paired dats is obtained from a single group of patients, unpaired data is obtained from two different groups of patients

Question 12

Describe the period prevalence

Answer 12

number of identified cases during a specified period of time / total number of people in that population

Question 13

describe the properties of normal distribution

Answer 13

68.3% of values lie within 1 SD of the mean
95.4% of values lie within 2 SD of the mean
99.7% of values lie within 3 SD of the mean

the standard deviation is a measure of how much dispersion exists from the mean

Question 14

describe the phases of clinical trials

Answer 14

phase 0 - exploratory studies
> small number of participants, aims to assess how drug behaves in human body
> assesses pharmacokinetics/pharmacodynamics

phase 1 - safety assessment
> determines side-effects prior to larger studies
> conducted on healthy volunteers

phase 2 - efficacy assessment
> small number of participants affected by particular disease
> IIa: optimal dosing, IIb: efficacy

phase 3 - effectiveness assessment
> 100-1000s of people, usually part of RCT
> compares new treatment with established treatment

phase 4 - postmarketing surveillance
> long-term effectiveness and side-effects