Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Alex's MRCP Cards > Stats > Flashcards

Stats Flashcards

1
Q

P value definition?

A

Probability of obtaining a result at least as extreme as the one that was actually observed, assuming that the null hypothesis is true

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

Type I and II errors?

A
  • Type I: the null hypothesis is rejected when it is true - i.e. Showing a difference between two groups when it doesn’t exist (= significance level)
  • Type II: the null hypothesis is accepted when it is false - i.e. Failing to spot a difference when
    one really exists
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Power?

A

Probability of (correctly) rejecting the null hypothesis when it is fale

  • Power = 1 - probability of a type II error
  • Can increase power by increasing sample size
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Types of significance test?

A

Parametric

  • Student’s t-test - paired or unpaired
  • Pearson’s product-moment coefficient - correlation

Non-Parametric

  • Mann-Whitney - unpaired data
  • Wilcoxon matched-pairs - compares two sets of observations on a single sample
  • Chi-squared test - used to compare proportions or percentages
  • Spearman, Kendall rank – correlation
  • McNemar’s test is used on nominal data to determine whether the row and column marginal
    frequencies are equal
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Correlation tests?

A

Parametric = Pearson’s coefficient

Non-parametric = Spearman’s coefficient

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

What is paired data?

A
  • Paired = data obtained from single group of patients
    • E.g. measurement before and after an intervention
  • Unpaired = data from two different groups of patients
    • Comparing response to different interventions in two groups
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Funnel plot?

A
  • Used to demonstrate publication bias in meta-analyses
  • Treatment effect (horizontal) study size (vertical)
  • Symmetrical inverted funnel shape indicates publication bias is unlikely, and vice versa
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Confidence intervals?

A
  • Describes the range of value around a mean, an odds ratio, a p value or a standard deviation within which the true value lies.
  • 95% CI –> 5% chance the true mean value for variable lies outside the range
  • CI = mean ± 2xSE (Standard Error)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

% of values within x SD of mean in a normal distribution?

A
  • 1 SD of mean = 68.3% of values
  • 2 SD of mean = 95.4% of values
  • 3 SD of mean = 99.7% of values

95% of values lie within 1.96 SD of the mean

NB:- 95.4% values wthin 2 SD of mean, leaving 4.6% outside the range. So 2.3% of values will be higher and 2.3% will be lower.

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

Standard deviation?

A

SD = square root of the variance

Average difference each observation in a sample lies from the sample mean

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

Mean, median mode in normal and skewed distributions?

A

Normal - mean = median = mode

  • Positively skewed (to the left) - mean > median > mode
  • Negatively skewed (to the right) - mean < median < mode

Alphabetical order

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

Standard error of the mean (SEM)?

A

Measure of spread expected for mean of the observations i.e. how accurate the calculated sample mean is from the true population mean

SEM = SD/square root(n)

So SEM gets smaller as n increases

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

Relative risk?

A

RR = Experimental event rate/Control event rate

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

Absolute risk reduction?

A

Absolute risk reduction = (Control event rate) - (Experimental event rate)

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

Relative risk reduction (RRR)?

A

RRR = (Absolute risk reduction)/(Control event rate)

i.e. RRR = (CER-EER)/CER

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

Number needed to treat?

A

Indicates how many patients would require an intervention to reduce the expected number of outcomes by 1.

NNT = 1/Absolute risk reduction

i.e. NNT = 1/(CER-EER)

17
Q

Odds/Odds ratio?

A

Odds = remember a ratio of the number of people who incur a particular outcome to the number of people who do not incur the outcome, NOT TO THE TOTAL NUMBER OF PEOPLE

Odds ratio = (Experimental odds)/(Control odds)

18
Q

Pre- and post-test odds and probability?

A
  1. Pre-test probability
    • Prevalence of a condition - proportion of people with the target disorder in the population at a specific time
  2. Post-test probability
    • Proportion of patients with that particular test result who have the target disorder
    • Post-test probability = post-test odds/(1 + post-test odds)
  3. Pre-test odds
    • The odds that the patient has the target disorder before the test is carried out
    • Pre-test odds = pre-test probability/(1 - pre-test probability)
  4. Post-test odds
    • The odds that the patient has the target disorder after the test is carried out
    • Post-test odds = pre-test odds x likelihood ratio

Likelihood ratio = sensitivity/(1-specificity)

19
Q

Incidence vs prevalence?

A
  1. Incidence
    • Number of new cases per population in a given time period
    • I.e. 40 new cases over past 12 months per 1000 population –> annual incidence is 0.04 or 4%
  2. Prevalence
    • Total number of cases per population at a particular point in time

Prevalence = incidence x duration of condition

Chronic diseases –> prevalence is much greater than the incidence

Acute diseases –> prevalence and incidence are similar

20
Q

Sensitivity and specificity?

A
  1. Sensitivity
    • Sensitive tests recognises sick people a sick - how many of the sick patients can the test identify by %?
    • High sensitivity –> negative test rules out
    • TP / (TP+FN)
  2. Specificity
    • Specific test recognises all healthy people as healthy - how many of the healthy patients can the test identify by %?
    • High specificity –> positive test rules in
    • Low type I error rate
    • TN / (TN+FP)
21
Q

Positive and negative predictive value?

A
  1. Positive predictive value
    • How many of the +ve test samples are actually sick
    • TP / (TP + FP)
  2. Negative predictive value
    • How many of the -ve test samples are actually healthy
    • TN / (TN + FN)

Both prevalence dependent

22
Q

Likelihood ratios?

A

Likelihood ratio for a positive test result

  • Sensitivity / (1-specificity)

Likelihood ratio for a negative test result

  • (1-sensitivity) / specificity

Remember to convert 80% to 0.8

Not prevalence dependent

23
Q

Wilson and Junger criteria for screening?

A
  1. The condition should be an important public health problem
  2. There should be an acceptable treatment for patients with recognised disease
  3. Facilities for diagnosis and treatment should be available
  4. There should be a recognised latent or early symptomatic stage
  5. The natural history of the condition, including its development from latent to declared disease should be adequately understood
  6. There should be a suitable test or examination
  7. The test or examination should be acceptable to the population
  8. There should be agreed policy on whom to treat
  9. The cost of case-finding (including diagnosis and subsequent treatment of patients) should be
    economically balanced in relation to the possible expenditure as a whole
  10. Case-finding should be a continuous process and not a ‘once and for all’ project
24
Q

Correlation?

A

Correlation co-efficient how closely the points lie to a line drawn through the plotted data

Denoted by value R - can lie between -1 and 1

  • R = 1 - strong positive correlation
  • R = 0 - no correlation
  • R = -1 - strong negative correlation
25
Q

Linear regression?

A

Can be used to predict how much one variable changes when a second variable is changed, which correlation cannot.

Y = A + BX

  • Y = variable being calculated
  • A = intercept value, when X = 0
  • B = slope of the line or regression co-efficient; how much Y changes for a given change in X
  • X = second variable
26
Q

Study design summary?

A
27
Q

Levels of evidence?

A
  • Ia - evidence from meta-analysis of randomised controlled trials
  • Ib - evidence from at least one randomised controlled trial
  • IIa - evidence from at least one well designed controlled trial which is not randomised
  • IIb - evidence from at least one well designed experimental trial
  • III - evidence from case, correlation and comparative studies
  • IV - evidence from a panel of experts
28
Q

Grading of recommendations

A
  • Grade A - based on evidence from at least one randomised controlled trial (i.e. Ia or Ib)
  • Grade B - based on evidence from non-randomised controlled trials (i.e. IIa, IIb or III)
  • Grade C - based on evidence from a panel of experts (i.e. IV)
29
Q

Intention to treat analysis?

A

All patients randomly assigned to one of the treatments are analysed together, regardless of whether or not they completed or received that treatment. Intention to treat analysis is done to avoid the effects of crossover and drop-out, which may affect the randomization to the treatment groups

30
Q

Study designs for new drugs (comparing to existing treatment)

A
  1. Superiority
    • Large sample size needed
  2. Equivalence
    • Equivalence margin defined (-delto to +delta) on a specified outcome
    • If the confidence interval of the difference between the two drugs lies within the equivalence margin then the drugs may be assumed to have a similar effect
  3. Non-inferiority
    • Similar to equivalence trials but only the lower confidence interval needs to lie within the equivalence margin (-delta)
    • Small sample sizes needed - could do larger superiority studies after non-inferiority proven

Alex's MRCP Cards (13 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions