Brainscape's Knowledge GenomeTM

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


See full index

oishi's final year flashcards > AFP Critical Appraisal & Stats > Flashcards

AFP Critical Appraisal & Stats Flashcards

1
Q

How will you summarise what the paper is about?

A 
Population = People with AF
Intervention = Apixaban 5mg BD
Control = Warfarin INR 2-3
Outcomes = Stroke or systemic embolism
Key findings
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What should you think about before summarising the paper?

A

QR

What is the research q and what is its relevance to clinical practise

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

What acronym do you use to assess internal validity?

A 
Recruitment
Allocation
Maintenance
Baseline
Outcome (and was it blinded)
Stats
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is internal validity?

A

The extent to which the observed results represent the truth in the population we are studying and, thus, are not due to methodological errors

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

How can we reduce selection bias?

A

Use consecutive recruitment rather than non-consecutive

Consider the recruitment location (e.g. GP vs hospital)

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

What types of recruitment are there?

A

Consecutive vs non-consecutive

Single centre vs multicentre

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

Types of allocation?

A

Randomised vs non randomised
Allocation blinded vs open label
Letters/Packets vs automated voice recognition system (electronic)
Block randomisation vs whole group randomisation
Cluster randomisation vs whole group randomisation

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

What is block randomisation?

A

Wait for X number of study participants and then say 50% go to each arm
Throughout the whole study there are equal numbers in both arms
Helps to mitigate temporal trends

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

What is cluster randomisation? Disadvantages?

A

Used in multicentre studies
Used when you can’t deliver different interventions in one place (e.g. poster in a GP practice)
Needs higher numbers of people
More difficult to achieve balance across groups

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

Benefit of randomisation?

A

Reduces risk of confounders

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

Why does blinding help?

A

Reduces risk of placebo effect

Reduces bias in data analysis

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

Benefit of Electronic voice recognition system?

A

Minimizes risk of tampering

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

What is study maintenance and what is usually aimed for?

A

Maintenance is the drop out rate from a study

Usually aim for <20% but ideally <10%

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

What bias is created if there is a poor maintenance?

A

Attrition bias - causes study to be under powered

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

Types of blinded outcome?

A

Single: patient only
Double: patient and physicians interacting
Triple: patient, physicians, outcome extractors

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

How can you blind if the outcome can’t be blinded e.g. surgical?

A

Use PROBE design
Prospective Randomised Open label Blinded endpoint adjudication
Whoever reviews the outcomes doesn’t know which arm of study the patient is in (e.g. radiologist when looking at stroke tx)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What do you need to look at in baseline?

A

Baseline characteristics of population and are they matched?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Besides blinding what else do you need to consider with outcome?

A

Is the choice of outcome important, adequate and important to patients?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Define power

A

the likelihood of detecting a difference when it exists (avoiding Type 2 error)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What 4 things should we look at when analysing the stats of a paper?

A

Power calculation and according recruitment target
Statistical models used for outcome measure (binary, continuous, time-dependent)
Effect size and significance level
Absolute risk reduction / Relative risk reduction / NNT

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What 3 things does power calculation depend on?

A
  1. How many events are expected over follow-up time (incidence x time) - increase sample size
  2. Expected improvement (RRR of 50%) - increase effect size
  3. What probability you want to detect the difference when it exists (80%?) - increase precision of measurement
    If a study is negative then think about the power!
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Define p value

A

probability that the association detected has arisen by chance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

How to strengthen the p value?

A

Size of association

Size of cohort

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What test do you use to calculate difference between 2 groups with categorical variables when not adjusted? E.g. stroke vs no stroke

A

Chi-squared

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

What test do you use to calculate difference between 2 groups with categorical variables when adjusted? E.g AF X Stroke

A

Binary logistic regression

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

What test do you use for comparing 2 groups with a continuous outcome e.g. Gender x BMI if normally distributed?

A

T-test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

What test do you use for comparing 2 groups with a continuous outcome e.g. Gender x BMI if not normally distributed?

A

Wilcoxon ranked test (dependent)/Mann Whitney U (independent)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

What test do you use for comparing >2 groups with a continuous outcome e.g. number of children x BMI if adjusted?

A

ANOVA

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

What test do you use for comparing >2 groups with a continuous outcome e.g. number of children x BMI if not adjusted?

A

ANCOVA

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

What test do you use to find a correlation between 2 parametric continuous variables?

A

Pearson’s rank

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

What test do you use to find a correlation between 2 non-parametric continuous variables?

A

Spearman’s rank

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
32
Q

What test to use for time to event data i.e. when time counts (e.g. mortality in cancer tx) to determine a direct association?

A

Kaplan Meier analysis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
33
Q

What test to use for time to event data i.e. when time counts (e.g. mortality in cancer tx) when adjusted for covariates?

A

Cox proportional hazards model

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
34
Q

How to assess external validity?

A

Resources - equipment/cost

Population - can you generalise the demographics

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
35
Q

What are the last 3 things we consider?

A

Funding - any conflicts of interest
Ethics - clinical equipoise, safety outcomes, data safety monitoring board
Conclusion

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
36
Q

Define 95% Confidence interval

A

A range, between which the population mean value will lie 95% of the time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
37
Q

Define relative risk

A

Risk of developing disease in the exposed group compared to the risk of developing disease in the unexposed group

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
38
Q

How do you calculate relative risk?

A

Those who got disease in exposed group/all exposed divided by those who got disease in non-exposed group/all non-exposed

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
39
Q

Which studies is relative risk used in?

A

Prospective cohort

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
40
Q

Define odds ratio?

A

Odds of something happening vs the odds of it not happening

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
41
Q

How do you calculate odds ratio?

A

Exposed with disease/not exposed with disease divided by exposed without disease/not exposed without disease

AKA odds in disease group/odds in control

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
42
Q

What studies is odds ratio used in?

A

Retrospective observational study (case-control study)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
43
Q

Define hazard ratio

A

Used to look at survival over time with a Kaplan Meier curve - equivalent to relative risk but risk is not constant with respect to time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
44
Q

What is a hazard ratio used in?

A

Prospective RCT

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
45
Q

Define incidence rate

A

Number of new cases over a defined period of time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
46
Q

Define prevalence

A

Number of cases in a given population at any given time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
47
Q

Define Absolute risk reduction (ARR) and how to calculate it

A

Difference in the incidence of disease between two groups

= P(event occurring in group 1) – P(event occurring in group 2)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
48
Q

Define NNT and how to calculate it

A

Number of patients who need to be treated to prevent one event occurring
= 1/ARR

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
49
Q

How to calculate RRR?

A

= ARR / P(event occurring in control group)

Express as %

50
Q

Define Type I error

A

Probability of rejecting H0 when in fact H0 is true
i.e. probability of concluding there is a significant difference when actually there’s no difference.
False positive

51
Q

Define Type II error

A

Probability of concluding H0 is true when in fact it is FALSE
i.e. false negative

52
Q

How do we set a type II error?

A

β is the probability of making type II error – usually set at 0.8

53
Q

How do we set a type I error?

A

Reflects p value cut-off i.e. usually = 0.05.

54
Q

What does per protocol analysis mean?
What is the benefit?
What is the con?

A

Only participants who complied with study protocol completely are included in analysis.
More accurate representation of treatment effect
Susceptible to attrition bias and exclusion bias

55
Q

What does intention to treat analysis involve?
What is the benefit?
What is the con?

A

All participants who have been randomised are included, regardless if they took the medicine / completed the study
More accurate representation of effect in clinical practice
Can’t determine what the drug does in optimal conditions

56
Q

What are case control studies used for?

A

Identifying associations between exposure and outcome

57
Q

Advantages of case control studies?

A 
Rare diseases
Quicker and cheaper to perform
Can analyse multiple exposures at once
Can calculate OR
Good for dynamic populations where follow up is difficult
58
Q

Disadvantages of case control studies?

A

Rely on quality of records
Selection bias and recall bias
Can’t demonstrate temporal association

59
Q

What is a cohort study used for?

A

Evidence for causation and temporal association

60
Q

Advantages of cohort studies?

A 
Assess temporality
Assess prognosis
Can control data collection and quality
Can calculate incidence – allows you to calculate RR, risk difference, NNT
Good for common exposures
61
Q

Disadvantages of cohort studies?

A 
Expensive 
Take longer
Not good for rare diseases
Loss to follow-up and potential of attrition bias
Can be affected by confounders
62
Q

Define bias

A

Factors which cause systematic over or under-estimate of a particular result

63
Q

What is selection bias?

A

Study sample does not represent entire population

64
Q

What is volunteer bias?

A

People who volunteer are different from population as a whole

65
Q

What is channelling bias?

A

Healthcare professionals may subconsciously select patients who would be good for study

66
Q

What is Interview bias?

A

If interviewer already knows disease status of participant

67
Q

What is recall bias?

A

Participants remember things differently

68
Q

What is Hawthorne bias?

A

Participants alter behaviour because they are aware of monitoring

69
Q

What is a confounder?

A

A factor which has an independent effect on both the exposure and the outcome

70
Q

How can we minimise confounders?

A

Stratification of participant groups
Regression analysis
Randomisation and matching

71
Q

What is external validity?

A

Generalisability of results

72
Q

Difference between RR and HR

A

RR looks at if an event has occurred during the study

HR looks at if an event occurred and when it occurred

73
Q

How to describe RR of 1.45?

A

45% more likely to have outcome X

74
Q

How to describe OR of 1.6?

A

Odds of exposure to factor X is 1.6x higher

75
Q

How to describe HR of 0.79?

A

At any particular point, group A were 21% less likely to have outcome X

76
Q

What is a type 1 error?

A 
Reject H0 (no difference) when actually H0 is true
I.e. False positive
77
Q

What is a type 2 error?

A

False negative

78
Q

Define sensitivity and how to calculate it

A

Probability of correctly identifying those with disease

TP/TP + FN

79
Q

Define specificity and how to calculate it

A

Probability of correctly identifying those without disease

TN/TN + FP

80
Q

What is a QALY? What is the accepted cost?

A

Quality adjusted life year

<20k per QALY – ACCEPT. 20-30 – think about

81
Q

What is a fixed effect meta analysis?

A

Assumes that studies are homogenous
All measuring same treatment effect
Calculate weighted average - give more weight to bigger studies

82
Q

What is a random effects meta?

A

Assumes heterogeneity between studies

83
Q

What methods are used to quantify heterogeneity?

A

Q – statistic - low p = heterogenetic

I2 – estimates proportion of total variance that is attributable to heterogeneity

84
Q

How is heterogeneity dealt with?

A

Sub group analysis – separate meta analyses

Meta regression - quantify how treatment effect changes with study characteristic

85
Q

What analysis approaches may be performed in an RCT?

A

Intention to treat
Per protocol
As treated

86
Q

What is a sensitivity analysis?

A

Assesses the impact of different assumptions and methodological choices on the results of the primary analysis
Can looks at differences in analysis approach, inclusion criteria, outcome definition, missing data
May also be good for missing data – can account for it using imputation

87
Q

What groups are responsible for overseeing a trial?

A

Trial management group – day to day
Data monitoring committee – independent – make recommendations. Perform formal interim analysis – stop trial if efficacy clear or serious adverse events
Trial steering committee – uses recommendation of DMC

88
Q

What are the issues with a formal interim analysis?

A

May stop trial prematurely due to random variation in treatment effect over time

89
Q

What is PPV?

A

Likelihood of people who test positive actually having disease
TP/TP + FP

90
Q

What is NPV?

A

Likelihood of people who test negative actually not having disease
TN/TN + FN

91
Q

What is heterogeneity?

A

the degree of difference between methodology in the studies and thus the treatment effect being measured

92
Q

What does asymmetry in a funnel plot (meta analysis) suggest and what does it lead to?

A 
publication bias (only studies that get good results are published)
Leads to overestimation of the treatment effect
93
Q

How do you overcome publication bias?

A

Register trial beforehand

Publish a protocol

94
Q

What is clinical equipoise?

A

a state of genuine uncertainty on the part of the clinical investigator regarding the comparative therapeutic merits of each arm in a trial

95
Q

What is Ecological fallacy and give an example

A

Assuming that the results of a cross sectional study relate to an individual. Eg areas with higher salt consumption are more at risk of heart attacks. Does amount of salt consumption actually increase likelihood in an individual?

96
Q

How can you graphically represent a meta analysis?

A

Forest plot

97
Q

What is the difference between a retrospective cohort study and a case control study?

A

case control takes patients based on outcome status whereas retrospective cohort doesn’t

98
Q

Disadvantage of retrospective cohort study?

A

Potential to find reverse causality

Recall and interviewer bias

99
Q

Disadvantage of using 2:1 allocation?

A

Undermines clinical equipoise and can cause therapeutic mis-estimation
Reduces power or requires 12% more participants
Affects internal validity

100
Q

Advantages of using 2:1 allocation?

A

In early phase trials to determine clinical utility
If a tx is particularly expensive
Need for additional safety info - good for adverse events

101
Q

How to interpret a confidence interval looking at the difference between 2 groups?

A

If it contains the value 0 then p>0.05

102
Q

How to interpret a confidence interval when comparing groups using a ratio instead of a difference e.g. odds/RR?

A

If it contains the value 1 then p>0.05

103
Q

Difference between OR and RR?

A

odds ratio is a ratio of two odds whereas the relative risk is a ratio of two probabilities

104
Q

Where can you get research funding from?

A 
Non-commercial:
Research councils - HRA, MRC
Government 
NIHR
Charities - CRUK

Commercial:
Pharma companies
Industry companies

105
Q

Where do you apply for ethical approval?

A

Health Research Authority Research Ethics Committee

106
Q

Which studies is odds ratio used in?

A

case control

107
Q

Which stats test to measure strength of association between 2 variables? (ND and NND)

A

ND: Pearson correlation
NND: Spearman rank

108
Q

Which descriptive stats for ND vs NND?

A

ND: Mean, SD
NND: Median, IQR

109
Q

Which stats test for comparison of one group to hypothetical value? ND, NND and categorical

A

ND: one sample T-test
NND: Wilcoxon
Categorical: Chi square/binomial

110
Q

Difference between one sample T-test and independent T-test? What is a paired T-test?

A

One sample compares a group to a pre-determined value
Independent compares a group to another group
Paired compares values within a group at different intervals

111
Q

Stats test for comparing 2 unpaired groups if ND, NND or categorical?

A

ND: Unpaired T-test
NND: Mann Whitney U
Categorical: Fisher’s exact test or Chi-squared if large groups

112
Q

Stats test for comparing 2 paired groups if ND, NND or categorical?

A

ND: Paired T-test
NND: Wilcoxon
Categorical: McNemar’s

113
Q

Stats test for comparing 3+ unmatched groups if ND, NND or categorical?

A

ND: one way ANOVA
NND: Kruskal Wallis
Categorical: Chi squared

114
Q

Stats test for comparing 3+ matched groups if ND, NND or categorical?

A

ND: Repeated measures ANOVA
NND: Friedman test
Categorical: Cochrane Q

115
Q

What stats test to use for prediction?

A

Single variable: Linear/logistic regression
Multiple variables: Multiple linear/logistic regressions

116
Q

What outcome does logistic regression provide?

A

Adjusted odds ratio

117
Q

What to use when the denominator is number of person-years?

A

Relative risk

118
Q

What is the advantage of using parametric tests over non-parametric?

A

Provides more statistical power

119
Q

What is the positive likelihood ratio?

A

The probability that a person with the disease tested positive for the disease divided by the probability that a person without the disease tested positive for the disease

AKA True positive / False positive

LR+ = sensitivity / (1 - specificity)

120
Q

What is the negative likelihood ratio?

A

False negatives / True negatives
LR- = (1 - sensitivity) / specificity

121
Q

Why is it important to do post hoc tests to investigate a significant ANOVA test?

A

They apply a correction for multiple testing to avoid a type 1 error

122
Q

What are likelihood ratios used for?

A

They assess the potential use of a diagnostic test and the likelihood of the patient having the disease

A ratio of 0-1 reduces chance of disease
Ratio >1 increases probability of having disease

oishi's final year flashcards (10 decks)

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